Artificial Neural Network Based State Estimators Integrated into Kalmtool
DEFF Research Database (Denmark)
Bayramoglu, Enis; Ravn, Ole; Poulsen, Niels Kjølstad
2012-01-01
In this paper we present a toolbox enabling easy evaluation and comparison of dierent ltering algorithms. The toolbox is called Kalmtool and is a set of MATLAB tools for state estimation of nonlinear systems. The toolbox now contains functions for Articial Neural Network Based State Estimation...
Design of Artificial Neural Network-Based pH Estimator
Directory of Open Access Journals (Sweden)
Shebel A. Alsabbah
2010-10-01
Full Text Available Taking into consideration the cost, size and drawbacks might be found with real hardware instrument for measuring pH values such that the complications of the wiring, installing, calibrating and troubleshooting the system, would make a person look for a cheaper, accurate, and alternative choice to perform the measuring operation, Where’s hereby, a feedforward artificial neural network-based pH estimator has to be proposed. The proposed estimator has been designed with multi- layer perceptrons. One input which is a measured base stream and two outputs represent pH values at strong base and strong/weak acids for a titration process. The created data base has been obtained with consideration of temperature variation. The final numerical results ensure the effectiveness and robustness of the design neural network-based pH estimator.
Estimation of tool wear during CNC milling using neural network-based sensor fusion
Ghosh, N.; Ravi, Y. B.; Patra, A.; Mukhopadhyay, S.; Paul, S.; Mohanty, A. R.; Chattopadhyay, A. B.
2007-01-01
Cutting tool wear degrades the product quality in manufacturing processes. Monitoring tool wear value online is therefore needed to prevent degradation in machining quality. Unfortunately there is no direct way of measuring the tool wear online. Therefore one has to adopt an indirect method wherein the tool wear is estimated from several sensors measuring related process variables. In this work, a neural network-based sensor fusion model has been developed for tool condition monitoring (TCM). Features extracted from a number of machining zone signals, namely cutting forces, spindle vibration, spindle current, and sound pressure level have been fused to estimate the average flank wear of the main cutting edge. Novel strategies such as, signal level segmentation for temporal registration, feature space filtering, outlier removal, and estimation space filtering have been proposed. The proposed approach has been validated by both laboratory and industrial implementations.
Zhang, Li
With the deregulation of the electric power market in New England, an independent system operator (ISO) has been separated from the New England Power Pool (NEPOOL). The ISO provides a regional spot market, with bids on various electricity-related products and services submitted by utilities and independent power producers. A utility can bid on the spot market and buy or sell electricity via bilateral transactions. Good estimation of market clearing prices (MCP) will help utilities and independent power producers determine bidding and transaction strategies with low risks, and this is crucial for utilities to compete in the deregulated environment. MCP prediction, however, is difficult since bidding strategies used by participants are complicated and MCP is a non-stationary process. The main objective of this research is to provide efficient short-term load and MCP forecasting and corresponding confidence interval estimation methodologies. In this research, the complexity of load and MCP with other factors is investigated, and neural networks are used to model the complex relationship between input and output. With improved learning algorithm and on-line update features for load forecasting, a neural network based load forecaster was developed, and has been in daily industry use since summer 1998 with good performance. MCP is volatile because of the complexity of market behaviors. In practice, neural network based MCP predictors usually have a cascaded structure, as several key input factors need to be estimated first. In this research, the uncertainties involved in a cascaded neural network structure for MCP prediction are analyzed, and prediction distribution under the Bayesian framework is developed. A fast algorithm to evaluate the confidence intervals by using the memoryless Quasi-Newton method is also developed. The traditional back-propagation algorithm for neural network learning needs to be improved since MCP is a non-stationary process. The extended Kalman
Zeng, Y; Zhang, J; Yin, H; Pan, Y
2007-01-01
Visual evoked potentials (VEPs) are time-varying signals typically buried in relatively large background noise known as the electroencephalogram (EEG). In this paper, an adaptive noise cancellation with neural network-based fuzzy inference system (NNFIS) was used and the NNFIS was carefully designed to model the VEP signal. It is assumed that VEP responses can be modelled by NNFIS with the centres of its membership functions evenly distributed over time. The weights of NNFIS are adaptively determined by minimizing the variance of the error signal using the least mean squares (LMS) algorithm. As the NNFIS is dynamic to any change of VEP, the non-stationary characteristics of VEP can be tracked. Thus, this method should be able to track the VEP. Four sets of simulated data indicate that the proposed method is appropriate to estimate VEP. A total of 150 trials are processed to demonstrate the superior performance of the proposed method.
Hour-Glass Neural Network Based Daily Money Flow Estimation for Automatic Teller Machines
Karungaru, Stephen; Akashi, Takuya; Nakano, Miyoko; Fukumi, Minoru
Monetary transactions using Automated Teller Machines (ATMs) have become a normal part of our daily lives. At ATMs, one can withdraw, send or debit money and even update passbooks among many other possible functions. ATMs are turning the banking sector into a ubiquitous service. However, while the advantages for the ATM users (financial institution customers) are many, the financial institution side faces an uphill task in management and maintaining the cash flow in the ATMs. On one hand, too much money in a rarely used ATM is wasteful, while on the other, insufficient amounts would adversely affect the customers and may result in a lost business opportunity for the financial institution. Therefore, in this paper, we propose a daily cash flow estimation system using neural networks that enables better daily forecasting of the money required at the ATMs. The neural network used in this work is a five layered hour glass shaped structure that achieves fast learning, even for the time series data for which seasonality and trend feature extraction is difficult. Feature extraction is carried out using the Akamatsu Integral and Differential transforms. This work achieves an average estimation accuracy of 92.6%.
Artificial neural network-based all-sky power estimation and fault detection in photovoltaic modules
Jazayeri, Kian; Jazayeri, Moein; Uysal, Sener
2017-04-01
The development of a system for output power estimation and fault detection in photovoltaic (PV) modules using an artificial neural network (ANN) is presented. Over 30,000 healthy and faulty data sets containing per-minute measurements of PV module output power (W) and irradiance (W/m2) along with real-time calculations of the Sun's position in the sky and the PV module surface temperature, collected during a three-month period, are fed to different ANNs as training paths. The first ANN being trained on healthy data is used for PV module output power estimation and the second ANN, which is trained on both healthy and faulty data, is utilized for PV module fault detection. The proposed PV module-level fault detection algorithm can expectedly be deployed in broader PV fleets by taking developmental considerations. The machine-learning-based automated system provides the possibility of all-sky real-time monitoring and fault detection of PV modules under any meteorological condition. Utilizing the proposed system, any power loss caused by damaged cells, shading conditions, accumulated dirt and dust on module surface, etc., is detected and reported immediately, potentially yielding increased reliability and efficiency of the PV systems and decreased support and maintenance costs.
Directory of Open Access Journals (Sweden)
Mahmut Ozan Gökkan
2017-01-01
Full Text Available Optical parameters (properties of tissue-mimicking phantoms are determined through noninvasive optical imaging. Objective of this study is to decompose obtained diffuse reflectance into these optical properties such as absorption and scattering coefficients. To do so, transmission spectroscopy is firstly used to measure the coefficients via an experimental setup. Next, the optical properties of each characterized phantom are input for Monte Carlo (MC simulations to get diffuse reflectance. Also, a surface image for each single phantom with its known optical properties is obliquely captured due to reflectance-based geometrical setup using CMOS camera that is positioned at 5∘ angle to the phantoms. For the illumination of light, a laser light source at 633nm wavelength is preferred, because optical properties of different components in a biological tissue on that wavelength are nonoverlapped. During in vitro measurements, we prepared 30 different mixture samples adding clinoleic intravenous lipid emulsion (CILE and evans blue (EB dye into a distilled water. Finally, all obtained diffuse reflectance values are used to estimate the optical coefficients by artificial neural networks (ANNs in inverse modeling. For a biological tissue it is found that the simulated and measured values in our results are in good agreement.
Wang, Yonggang; Li, Deng; Lu, Xiaoming; Cheng, Xinyi; Wang, Liwei
2014-10-01
Continuous crystal-based positron emission tomography (PET) detectors could be an ideal alternative for current high-resolution pixelated PET detectors if the issues of high performance γ interaction position estimation and its real-time implementation are solved. Unfortunately, existing position estimators are not very feasible for implementation on field-programmable gate array (FPGA). In this paper, we propose a new self-organizing map neural network-based nearest neighbor (SOM-NN) positioning scheme aiming not only at providing high performance, but also at being realistic for FPGA implementation. Benefitting from the SOM feature mapping mechanism, the large set of input reference events at each calibration position is approximated by a small set of prototypes, and the computation of the nearest neighbor searching for unknown events is largely reduced. Using our experimental data, the scheme was evaluated, optimized and compared with the smoothed k-NN method. The spatial resolutions of full-width-at-half-maximum (FWHM) of both methods averaged over the center axis of the detector were obtained as 1.87 ±0.17 mm and 1.92 ±0.09 mm, respectively. The test results show that the SOM-NN scheme has an equivalent positioning performance with the smoothed k-NN method, but the amount of computation is only about one-tenth of the smoothed k-NN method. In addition, the algorithm structure of the SOM-NN scheme is more feasible for implementation on FPGA. It has the potential to realize real-time position estimation on an FPGA with a high-event processing throughput.
DEFF Research Database (Denmark)
Soliman, Hammam Abdelaal Hammam; Abdelsalam, Ibrahim; Wang, Huai
2017-01-01
, a proposed software condition monitoring methodology based on Artificial Neural Network (ANN) algorithm is presented. Matlab software is used to train and generate the proposed ANN. The proposed methodology estimates the capacitance of the DC-link capacitor in a three phase front-end diode bridge AC......In modern design of power electronic converters, reliability of DC-link capacitors is an essential aspect to be considered. The industrial field have been attracted to the monitoring of their health condition and the estimation of their ageing process status. The existing condition monitoring...
Ruiz, Jonatan R; Ramirez-Lechuga, Jorge; Ortega, Francisco B; Castro-Piñero, José; Benitez, Jose M; Arauzo-Azofra, Antonio; Sanchez, Cristobal; Sjöström, Michael; Castillo, Manuel J; Gutierrez, Angel; Zabala, Mikel
2008-11-01
To develop an artificial neural network (ANN)-equation to estimate maximal oxygen uptake (VO(2max)) from 20m shuttle run test (20 mSRT) performance (stage), sex, age, weight, and height in young persons. The 20 mSRT was performed by 193 (122 boys and 71 girls) adolescents aged 13-19 years. All the adolescents wore a portable gas analyzer to measure VO(2) and heart rate during the test. The equation was developed and cross-validated following the ANN mathematical model. The neural net performance was assessed through several error measures. Agreement between the measured VO(2max) and estimated VO(2max) from Léger's and ANN equations were analysed following the Bland and Altman method. The percentage error was 17.13 and 7.38 for Léger and ANN-equation (P<0.001), respectively, and the standard error of the estimate obtained with Léger's equation was 4.27 ml/(kg min), while for the ANN-equation was 2.84 ml/(kg min). A Bland-Altman plot for the measured VO(2max) and Léger-VO(2max) showed a mean difference of 4.9 ml/(kg min) (P<0.001), while the Bland-Altman plot for the measured VO(2max) and ANN-VO(2max) showed a mean difference of 0.5 ml/(kg min) (P=0.654). In the validation sample, the percentage error was 21.08 and 8.68 for Léger and ANN-equation (P<0.001), respectively. In this study, an ANN-based equation to estimate VO(2max) from 20 mSRT performance (stage), sex, age, weight, and height in adolescents was developed and cross-validated. The newly developed equation was shown to be more accurate than Léger's. The proposed model has been coded in a user-friendly spreadsheet.
DEFF Research Database (Denmark)
Landschützer, P.; Gruber, N.; Bakker, D.C.E.
2013-01-01
pressure of CO2 (pCO2) at a resolution of 1° × 1°. From those, we compute the air–sea CO2 flux maps using a standard gas exchange parameterization and high-resolution wind speeds. The neural networks fit the observed pCO2 data with a root mean square error (RMSE) of about 10 μatm and with almost no bias......) a continuous improvement of the observations, i.e., the Surface Ocean CO2 Atlas (SOCAT) v1.5 database and (ii) a newly developed technique to interpolate the observations in space and time. In particular, we use a 2 step neural network approach to reconstruct basin-wide monthly maps of the sea surface partial...
Low-dimensional recurrent neural network-based Kalman filter for speech enhancement.
Xia, Youshen; Wang, Jun
2015-07-01
This paper proposes a new recurrent neural network-based Kalman filter for speech enhancement, based on a noise-constrained least squares estimate. The parameters of speech signal modeled as autoregressive process are first estimated by using the proposed recurrent neural network and the speech signal is then recovered from Kalman filtering. The proposed recurrent neural network is globally asymptomatically stable to the noise-constrained estimate. Because the noise-constrained estimate has a robust performance against non-Gaussian noise, the proposed recurrent neural network-based speech enhancement algorithm can minimize the estimation error of Kalman filter parameters in non-Gaussian noise. Furthermore, having a low-dimensional model feature, the proposed neural network-based speech enhancement algorithm has a much faster speed than two existing recurrent neural networks-based speech enhancement algorithms. Simulation results show that the proposed recurrent neural network-based speech enhancement algorithm can produce a good performance with fast computation and noise reduction. Copyright © 2015 Elsevier Ltd. All rights reserved.
Neural network based system for equipment surveillance
Vilim, R.B.; Gross, K.C.; Wegerich, S.W.
1998-04-28
A method and system are disclosed for performing surveillance of transient signals of an industrial device to ascertain the operating state. The method and system involves the steps of reading into a memory training data, determining neural network weighting values until achieving target outputs close to the neural network output. If the target outputs are inadequate, wavelet parameters are determined to yield neural network outputs close to the desired set of target outputs and then providing signals characteristic of an industrial process and comparing the neural network output to the industrial process signals to evaluate the operating state of the industrial process. 33 figs.
A Neural Network-Based Interval Pattern Matcher
Directory of Open Access Journals (Sweden)
Jing Lu
2015-07-01
Full Text Available One of the most important roles in the machine learning area is to classify, and neural networks are very important classifiers. However, traditional neural networks cannot identify intervals, let alone classify them. To improve their identification ability, we propose a neural network-based interval matcher in our paper. After summarizing the theoretical construction of the model, we take a simple and a practical weather forecasting experiment, which show that the recognizer accuracy reaches 100% and that is promising.
Representations in neural network based empirical potentials
Cubuk, Ekin D.; Malone, Brad D.; Onat, Berk; Waterland, Amos; Kaxiras, Efthimios
2017-07-01
Many structural and mechanical properties of crystals, glasses, and biological macromolecules can be modeled from the local interactions between atoms. These interactions ultimately derive from the quantum nature of electrons, which can be prohibitively expensive to simulate. Machine learning has the potential to revolutionize materials modeling due to its ability to efficiently approximate complex functions. For example, neural networks can be trained to reproduce results of density functional theory calculations at a much lower cost. However, how neural networks reach their predictions is not well understood, which has led to them being used as a "black box" tool. This lack of understanding is not desirable especially for applications of neural networks in scientific inquiry. We argue that machine learning models trained on physical systems can be used as more than just approximations since they had to "learn" physical concepts in order to reproduce the labels they were trained on. We use dimensionality reduction techniques to study in detail the representation of silicon atoms at different stages in a neural network, which provides insight into how a neural network learns to model atomic interactions.
Optical-Correlator Neural Network Based On Neocognitron
Chao, Tien-Hsin; Stoner, William W.
1994-01-01
Multichannel optical correlator implements shift-invariant, high-discrimination pattern-recognizing neural network based on paradigm of neocognitron. Selected as basic building block of this neural network because invariance under shifts is inherent advantage of Fourier optics included in optical correlators in general. Neocognitron is conceptual electronic neural-network model for recognition of visual patterns. Multilayer processing achieved by iteratively feeding back output of feature correlator to input spatial light modulator and updating Fourier filters. Neural network trained by use of characteristic features extracted from target images. Multichannel implementation enables parallel processing of large number of selected features.
Analog neural network-based helicopter gearbox health monitoring system.
Monsen, P T; Dzwonczyk, M; Manolakos, E S
1995-12-01
The development of a reliable helicopter gearbox health monitoring system (HMS) has been the subject of considerable research over the past 15 years. The deployment of such a system could lead to a significant saving in lives and vehicles as well as dramatically reduce the cost of helicopter maintenance. Recent research results indicate that a neural network-based system could provide a viable solution to the problem. This paper presents two neural network-based realizations of an HMS system. A hybrid (digital/analog) neural system is proposed as an extremely accurate off-line monitoring tool used to reduce helicopter gearbox maintenance costs. In addition, an all analog neural network is proposed as a real-time helicopter gearbox fault monitor that can exploit the ability of an analog neural network to directly compute the discrete Fourier transform (DFT) as a sum of weighted samples. Hardware performance results are obtained using the Integrated Neural Computing Architecture (INCA/1) analog neural network platform that was designed and developed at The Charles Stark Draper Laboratory. The results indicate that it is possible to achieve a 100% fault detection rate with 0% false alarm rate by performing a DFT directly on the first layer of INCA/1 followed by a small-size two-layer feed-forward neural network and a simple post-processing majority voting stage.
Neural Network Based Intelligent Sootblowing System
Energy Technology Data Exchange (ETDEWEB)
Mark Rhode
2005-04-01
. Due to the composition of coal, particulate matter is also a by-product of coal combustion. Modern day utility boilers are usually fitted with electrostatic precipitators to aid in the collection of particulate matter. Although extremely efficient, these devices are sensitive to rapid changes in inlet mass concentration as well as total mass loading. Traditionally, utility boilers are equipped with devices known as sootblowers, which use, steam, water or air to dislodge and clean the surfaces within the boiler and are operated based upon established rule or operator's judgment. Poor sootblowing regimes can influence particulate mass loading to the electrostatic precipitators. The project applied a neural network intelligent sootblowing system in conjunction with state-of-the-art controls and instruments to optimize the operation of a utility boiler and systematically control boiler slagging/fouling. This optimization process targeted reduction of NOx of 30%, improved efficiency of 2% and a reduction in opacity of 5%. The neural network system proved to be a non-invasive system which can readily be adapted to virtually any utility boiler. Specific conclusions from this neural network application are listed below. These conclusions should be used in conjunction with the specific details provided in the technical discussions of this report to develop a thorough understanding of the process.
Advanced neural network-based computational schemes for robust fault diagnosis
Mrugalski, Marcin
2014-01-01
The present book is devoted to problems of adaptation of artificial neural networks to robust fault diagnosis schemes. It presents neural networks-based modelling and estimation techniques used for designing robust fault diagnosis schemes for non-linear dynamic systems. A part of the book focuses on fundamental issues such as architectures of dynamic neural networks, methods for designing of neural networks and fault diagnosis schemes as well as the importance of robustness. The book is of a tutorial value and can be perceived as a good starting point for the new-comers to this field. The book is also devoted to advanced schemes of description of neural model uncertainty. In particular, the methods of computation of neural networks uncertainty with robust parameter estimation are presented. Moreover, a novel approach for system identification with the state-space GMDH neural network is delivered. All the concepts described in this book are illustrated by both simple academic illustrative examples and practica...
Neural network based data-driven predictor: Case study on clinker ...
African Journals Online (AJOL)
Soft sensors are key solutions in process industries. Important parameters which are difficult or cost a lot to measure can be predicted using soft sensors. In this paper neural network based clinker quality predictor is developed. The predictor genuinely estimates LSF, SM, AM and C3S values. There is a time delay while ...
Acharya, Nachiketa; Shrivastava, Nitin Anand; Panigrahi, B. K.; Mohanty, U. C.
2014-09-01
The south peninsular part of India gets maximum amount of rainfall during the northeast monsoon (NEM) season [October to November (OND)] which is the primary source of water for the agricultural activities in this region. A nonlinear method viz., Extreme learning machine (ELM) has been employed on general circulation model (GCM) products to make the multi-model ensemble (MME) based estimation of NEM rainfall (NEMR). The ELM is basically is an improved learning algorithm for the single feed-forward neural network (SLFN) architecture. The 27 year (1982-2008) lead-1 (using initial conditions of September for forecasting the mean rainfall of OND) hindcast runs (1982-2008) from seven GCM has been used to make MME. The improvement of the proposed method with respect to other regular MME (simple arithmetic mean of GCMs (EM) and singular value decomposition based multiple linear regressions based MME) has been assessed through several skill metrics like Spread distribution, multiplicative bias, prediction errors, the yield of prediction, Pearson's and Kendal's correlation coefficient and Wilmort's index of agreement. The efficiency of ELM estimated rainfall is established by all the stated skill scores. The performance of ELM in extreme NEMR years, out of which 4 years are characterized by deficit rainfall and 5 years are identified as excess, is also examined. It is found that the ELM could expeditiously capture these extremes reasonably well as compared to the other MME approaches.
Wang, Jeff; Kato, Fumi; Yamashita, Hiroko; Baba, Motoi; Cui, Yi; Li, Ruijiang; Oyama-Manabe, Noriko; Shirato, Hiroki
2017-04-01
Breast cancer is the most common invasive cancer among women and its incidence is increasing. Risk assessment is valuable and recent methods are incorporating novel biomarkers such as mammographic density. Artificial neural networks (ANN) are adaptive algorithms capable of performing pattern-to-pattern learning and are well suited for medical applications. They are potentially useful for calibrating full-field digital mammography (FFDM) for quantitative analysis. This study uses ANN modeling to estimate volumetric breast density (VBD) from FFDM on Japanese women with and without breast cancer. ANN calibration of VBD was performed using phantom data for one FFDM system. Mammograms of 46 Japanese women diagnosed with invasive carcinoma and 53 with negative findings were analyzed using ANN models learned. ANN-estimated VBD was validated against phantom data, compared intra-patient, with qualitative composition scoring, with MRI VBD, and inter-patient with classical risk factors of breast cancer as well as cancer status. Phantom validations reached an R 2 of 0.993. Intra-patient validations ranged from R 2 of 0.789 with VBD to 0.908 with breast volume. ANN VBD agreed well with BI-RADS scoring and MRI VBD with R 2 ranging from 0.665 with VBD to 0.852 with breast volume. VBD was significantly higher in women with cancer. Associations with age, BMI, menopause, and cancer status previously reported were also confirmed. ANN modeling appears to produce reasonable measures of mammographic density validated with phantoms, with existing measures of breast density, and with classical biomarkers of breast cancer. FFDM VBD is significantly higher in Japanese women with cancer.
Chang, C.-H.; Johnson, N. C.; Cassar, N.
2014-06-01
Southern Ocean organic carbon export plays an important role in the global carbon cycle, yet its basin-scale climatology and variability are uncertain due to limited coverage of in situ observations. In this study, a neural network approach based on the self-organizing map (SOM) is adopted to construct weekly gridded (1° × 1°) maps of organic carbon export for the Southern Ocean from 1998 to 2009. The SOM is trained with in situ measurements of O2 / Ar-derived net community production (NCP) that are tightly linked to the carbon export in the mixed layer on timescales of one to two weeks and with six potential NCP predictors: photosynthetically available radiation (PAR), particulate organic carbon (POC), chlorophyll (Chl), sea surface temperature (SST), sea surface height (SSH), and mixed layer depth (MLD). This nonparametric approach is based entirely on the observed statistical relationships between NCP and the predictors and, therefore, is strongly constrained by observations. A thorough cross-validation yields three retained NCP predictors, Chl, PAR, and MLD. Our constructed NCP is further validated by good agreement with previously published, independent in situ derived NCP of weekly or longer temporal resolution through real-time and climatological comparisons at various sampling sites. The resulting November-March NCP climatology reveals a pronounced zonal band of high NCP roughly following the Subtropical Front in the Atlantic, Indian, and western Pacific sectors, and turns southeastward shortly after the dateline. Other regions of elevated NCP include the upwelling zones off Chile and Namibia, the Patagonian Shelf, the Antarctic coast, and areas surrounding the Islands of Kerguelen, South Georgia, and Crozet. This basin-scale NCP climatology closely resembles that of the satellite POC field and observed air-sea CO2 flux. The long-term mean area-integrated NCP south of 50° S from our dataset, 17.9 mmol C m-2 d-1, falls within the range of 8.3 to 24 mmol
Chang, C.-H.; Johnson, N. C.; Cassar, N.
2013-10-01
Southern Ocean organic carbon export plays an important role in the global carbon cycle, yet its basin-scale climatology and variability are uncertain due to limited coverage of in situ observations. In this study, a neural network approach based on the self-organizing map (SOM) is adopted to construct weekly gridded (1° × 1°) maps of organic carbon export for the Southern Ocean from 1998 to 2009. The SOM is trained with in situ measurements of O2 / Ar-derived net community production (NCP) that are tightly linked to the carbon export in the mixed layer on timescales of 1-2 weeks, and six potential NCP predictors: photosynthetically available radiation (PAR), particulate organic carbon (POC), chlorophyll (Chl), sea surface temperature (SST), sea surface height (SSH), and mixed layer depth (MLD). This non-parametric approach is based entirely on the observed statistical relationships between NCP and the predictors, and therefore is strongly constrained by observations. A thorough cross-validation yields three retained NCP predictors, Chl, PAR, and MLD. Our constructed NCP is further validated by good agreement with previously published independent in situ derived NCP of weekly or longer temporal resolution through real-time and climatological comparisons at various sampling sites. The resulting November-March NCP climatology reveals a pronounced zonal band of high NCP roughly following the subtropical front in the Atlantic, Indian and western Pacific sectors, and turns southeastward shortly after the dateline. Other regions of elevated NCP include the upwelling zones off Chile and Namibia, Patagonian Shelf, Antarctic coast, and areas surrounding the Islands of Kerguelen, South Georgia, and Crozet. This basin-scale NCP climatology closely resembles that of the satellite POC field and observed air-sea CO2 flux. The long-term mean area-integrated NCP south of 50° S from our dataset, 14 mmol C m-2 d-1, falls within the range of 8.3-24 mmol C m-2 d-1 from other model
Neural network based feed-forward high density associative memory
Daud, T.; Moopenn, A.; Lamb, J. L.; Ramesham, R.; Thakoor, A. P.
1987-01-01
A novel thin film approach to neural-network-based high-density associative memory is described. The information is stored locally in a memory matrix of passive, nonvolatile, binary connection elements with a potential to achieve a storage density of 10 to the 9th bits/sq cm. Microswitches based on memory switching in thin film hydrogenated amorphous silicon, and alternatively in manganese oxide, have been used as programmable read-only memory elements. Low-energy switching has been ascertained in both these materials. Fabrication and testing of memory matrix is described. High-speed associative recall approaching 10 to the 7th bits/sec and high storage capacity in such a connection matrix memory system is also described.
Neural Network Based Intrusion Detection System for Critical Infrastructures
Energy Technology Data Exchange (ETDEWEB)
Todd Vollmer; Ondrej Linda; Milos Manic
2009-07-01
Resiliency and security in control systems such as SCADA and Nuclear plant’s in today’s world of hackers and malware are a relevant concern. Computer systems used within critical infrastructures to control physical functions are not immune to the threat of cyber attacks and may be potentially vulnerable. Tailoring an intrusion detection system to the specifics of critical infrastructures can significantly improve the security of such systems. The IDS-NNM – Intrusion Detection System using Neural Network based Modeling, is presented in this paper. The main contributions of this work are: 1) the use and analyses of real network data (data recorded from an existing critical infrastructure); 2) the development of a specific window based feature extraction technique; 3) the construction of training dataset using randomly generated intrusion vectors; 4) the use of a combination of two neural network learning algorithms – the Error-Back Propagation and Levenberg-Marquardt, for normal behavior modeling. The presented algorithm was evaluated on previously unseen network data. The IDS-NNM algorithm proved to be capable of capturing all intrusion attempts presented in the network communication while not generating any false alerts.
An efficient neural network based method for medical image segmentation.
Torbati, Nima; Ayatollahi, Ahmad; Kermani, Ali
2014-01-01
The aim of this research is to propose a new neural network based method for medical image segmentation. Firstly, a modified self-organizing map (SOM) network, named moving average SOM (MA-SOM), is utilized to segment medical images. After the initial segmentation stage, a merging process is designed to connect the objects of a joint cluster together. A two-dimensional (2D) discrete wavelet transform (DWT) is used to build the input feature space of the network. The experimental results show that MA-SOM is robust to noise and it determines the input image pattern properly. The segmentation results of breast ultrasound images (BUS) demonstrate that there is a significant correlation between the tumor region selected by a physician and the tumor region segmented by our proposed method. In addition, the proposed method segments X-ray computerized tomography (CT) and magnetic resonance (MR) head images much better than the incremental supervised neural network (ISNN) and SOM-based methods. © 2013 Published by Elsevier Ltd.
Directory of Open Access Journals (Sweden)
Youtao Gao
2017-01-01
Full Text Available In order to improve the accuracy of the dynamical model used in the orbit determination of the Lagrangian navigation satellites, the nonlinear perturbations acting on Lagrangian navigation satellites are estimated by a neural network. A neural network based state observer is applied to autonomously determine the orbits of Lagrangian navigation satellites using only satellite-to-satellite range. This autonomous orbit determination method does not require linearizing the dynamical mode. There is no need to calculate the transition matrix. It is proved that three satellite-to-satellite ranges are needed using this method; therefore, the navigation constellation should include four Lagrangian navigation satellites at least. Four satellites orbiting on the collinear libration orbits are chosen to construct a constellation which is used to demonstrate the utility of this method. Simulation results illustrate that the stable error of autonomous orbit determination is about 10 m. The perturbation can be estimated by the neural network.
Sidelnikov, O. S.; Redyuk, A. A.; Sygletos, S.
2017-12-01
We consider neural network-based schemes of digital signal processing. It is shown that the use of a dynamic neural network-based scheme of signal processing ensures an increase in the optical signal transmission quality in comparison with that provided by other methods for nonlinear distortion compensation.
DEFF Research Database (Denmark)
Jørgensen, Ivan Harald Holger; Bogason, Gudmundur; Bruun, Erik
1995-01-01
is implemented using switched-current technique and is capable of estimating flow in the μl/s range. The neural estimator is built around a multiplierless neural network, containing 96 synaptic weights which are updated using the LMS1-algorithm. An experimental chip has been designed that operates at 5 V......This paper proposes a new way to estimate the flow in a micromechanical flow channel. A neural network is used to estimate the delay of random temperature fluctuations induced in a fluid. The design and implementation of a hardware efficient neural flow estimator is described. The system...
Boosting feature selection for Neural Network based regression.
Bailly, Kevin; Milgram, Maurice
2009-01-01
The head pose estimation problem is well known to be a challenging task in computer vision and is a useful tool for several applications involving human-computer interaction. This problem can be stated as a regression one where the input is an image and the output is pan and tilt angles. Finding the optimal regression is a hard problem because of the high dimensionality of the input (number of image pixels) and the large variety of morphologies and illumination. We propose a new method combining a boosting strategy for feature selection and a neural network for the regression. Potential features are a very large set of Haar-like wavelets which are well known to be adapted to face image processing. To achieve the feature selection, a new Fuzzy Functional Criterion (FFC) is introduced which is able to evaluate the link between a feature and the output without any estimation of the joint probability density function as in the Mutual Information. The boosting strategy uses this criterion at each step: features are evaluated by the FFC using weights on examples computed from the error produced by the neural network trained at the previous step. Tests are carried out on the commonly used Pointing 04 database and compared with three state-of-the-art methods. We also evaluate the accuracy of the estimation on FacePix, a database with a high angular resolution. Our method is compared positively to a Convolutional Neural Network, which is well known to incorporate feature extraction in its first layers.
Neural network based adaptive output feedback control: Applications and improvements
Kutay, Ali Turker
Application of recently developed neural network based adaptive output feedback controllers to a diverse range of problems both in simulations and experiments is investigated in this thesis. The purpose is to evaluate the theory behind the development of these controllers numerically and experimentally, identify the needs for further development in practical applications, and to conduct further research in directions that are identified to ultimately enhance applicability of adaptive controllers to real world problems. We mainly focus our attention on adaptive controllers that augment existing fixed gain controllers. A recently developed approach holds great potential for successful implementations on real world applications due to its applicability to systems with minimal information concerning the plant model and the existing controller. In this thesis the formulation is extended to the multi-input multi-output case for distributed control of interconnected systems and successfully tested on a formation flight wind tunnel experiment. The command hedging method is formulated for the approach to further broaden the class of systems it can address by including systems with input nonlinearities. Also a formulation is adopted that allows the approach to be applied to non-minimum phase systems for which non-minimum phase characteristics are modeled with sufficient accuracy and treated properly in the design of the existing controller. It is shown that the approach can also be applied to augment nonlinear controllers under certain conditions and an example is presented where the nonlinear guidance law of a spinning projectile is augmented. Simulation results on a high fidelity 6 degrees-of-freedom nonlinear simulation code are presented. The thesis also presents a preliminary adaptive controller design for closed loop flight control with active flow actuators. Behavior of such actuators in dynamic flight conditions is not known. To test the adaptive controller design in
Selected Flight Test Results for Online Learning Neural Network-Based Flight Control System
Williams-Hayes, Peggy S.
2004-01-01
The NASA F-15 Intelligent Flight Control System project team developed a series of flight control concepts designed to demonstrate neural network-based adaptive controller benefits, with the objective to develop and flight-test control systems using neural network technology to optimize aircraft performance under nominal conditions and stabilize the aircraft under failure conditions. This report presents flight-test results for an adaptive controller using stability and control derivative values from an online learning neural network. A dynamic cell structure neural network is used in conjunction with a real-time parameter identification algorithm to estimate aerodynamic stability and control derivative increments to baseline aerodynamic derivatives in flight. This open-loop flight test set was performed in preparation for a future phase in which the learning neural network and parameter identification algorithm output would provide the flight controller with aerodynamic stability and control derivative updates in near real time. Two flight maneuvers are analyzed - pitch frequency sweep and automated flight-test maneuver designed to optimally excite the parameter identification algorithm in all axes. Frequency responses generated from flight data are compared to those obtained from nonlinear simulation runs. Flight data examination shows that addition of flight-identified aerodynamic derivative increments into the simulation improved aircraft pitch handling qualities.
Structure Learning for Deep Neural Networks Based on Multiobjective Optimization.
Liu, Jia; Gong, Maoguo; Miao, Qiguang; Wang, Xiaogang; Li, Hao
2017-05-05
This paper focuses on the connecting structure of deep neural networks and proposes a layerwise structure learning method based on multiobjective optimization. A model with better generalization can be obtained by reducing the connecting parameters in deep networks. The aim is to find the optimal structure with high representation ability and better generalization for each layer. Then, the visible data are modeled with respect to structure based on the products of experts. In order to mitigate the difficulty of estimating the denominator in PoE, the denominator is simplified and taken as another objective, i.e., the connecting sparsity. Moreover, for the consideration of the contradictory nature between the representation ability and the network connecting sparsity, the multiobjective model is established. An improved multiobjective evolutionary algorithm is used to solve this model. Two tricks are designed to decrease the computational cost according to the properties of input data. The experiments on single-layer level, hierarchical level, and application level demonstrate the effectiveness of the proposed algorithm, and the learned structures can improve the performance of deep neural networks.
neural network based load frequency control for restructuring power
African Journals Online (AJOL)
2012-03-01
Mar 1, 2012 ... Abstract. In this study, an artificial neural network (ANN) application of load frequency control. (LFC) of a Multi-Area power system by using a neural network controller is presented. The comparison between a conventional Proportional Integral (PI) controller and the proposed artificial neural networks ...
Artificial neural Network-Based modeling and monitoring of photovoltaic generator
Directory of Open Access Journals (Sweden)
H. MEKKI
2015-03-01
Full Text Available In this paper, an artificial neural network based-model (ANNBM is introduced for partial shading detection losses in photovoltaic (PV panel. A Multilayer Perceptron (MLP is used to estimate the electrical outputs (current and voltage of the photovoltaic module using the external meteorological data: solar irradiation G (W/m2 and the module temperature T (°C. Firstly, a database of the BP150SX photovoltaic module operating without any defect has been used to train the considered MLP. Subsequently, in the first case of this study, the developed model is used to estimate the output current and voltage of the PV module considering the partial shading effect. Results confirm the good ability of the ANNBM to detect the partial shading effect in the photovoltaic module with logical accuracy. The proposed strategy could also be used for the online monitoring and supervision of PV modules.
Neural Network Based Load Frequency Control for Restructuring ...
African Journals Online (AJOL)
Electric load variations can happen independently in both units. Both neural controllers are trained with the back propagation-through-time algorithm. Use of a neural network to model the dynamic system is avoided by introducing the Jacobian matrices of the system in the back propagation chain used in controller training.
Implementation of neural network based non-linear predictive control
DEFF Research Database (Denmark)
Sørensen, Paul Haase; Nørgård, Peter Magnus; Ravn, Ole
1999-01-01
of non-linear systems. GPC is model based and in this paper we propose the use of a neural network for the modeling of the system. Based on the neural network model, a controller with extended control horizon is developed and the implementation issues are discussed, with particular emphasis...
A neural network based approach to social touch classification
van Wingerden, Siewart; Uebbing, Tobias J.; Jung, Merel Madeleine; Poel, Mannes
2014-01-01
Touch is an important interaction modality in social interaction, for instance touch can communicate emotions and can intensify emotions communicated by other modalities. In this paper we explore the use of Neural Networks for the classification of touch. The exploration and assessment of Neural
Neural network based adaptive control for nonlinear dynamic regimes
Shin, Yoonghyun
Adaptive control designs using neural networks (NNs) based on dynamic inversion are investigated for aerospace vehicles which are operated at highly nonlinear dynamic regimes. NNs play a key role as the principal element of adaptation to approximately cancel the effect of inversion error, which subsequently improves robustness to parametric uncertainty and unmodeled dynamics in nonlinear regimes. An adaptive control scheme previously named 'composite model reference adaptive control' is further developed so that it can be applied to multi-input multi-output output feedback dynamic inversion. It can have adaptive elements in both the dynamic compensator (linear controller) part and/or in the conventional adaptive controller part, also utilizing state estimation information for NN adaptation. This methodology has more flexibility and thus hopefully greater potential than conventional adaptive designs for adaptive flight control in highly nonlinear flight regimes. The stability of the control system is proved through Lyapunov theorems, and validated with simulations. The control designs in this thesis also include the use of 'pseudo-control hedging' techniques which are introduced to prevent the NNs from attempting to adapt to various actuation nonlinearities such as actuator position and rate saturations. Control allocation is introduced for the case of redundant control effectors including thrust vectoring nozzles. A thorough comparison study of conventional and NN-based adaptive designs for a system under a limit cycle, wing-rock, is included in this research, and the NN-based adaptive control designs demonstrate their performances for two highly maneuverable aerial vehicles, NASA F-15 ACTIVE and FQM-117B unmanned aerial vehicle (UAV), operated under various nonlinearities and uncertainties.
Neural Network-Based Segmentation of Textures Using Gabor Features
Ramakrishnan, AG; Raja, Kumar S; Ram, Ragu HV
2002-01-01
The effectiveness of Gabor filters for texture segmentation is well known. In this paper, we propose a texture identification scheme, based on a neural network (NN) using Gabor features. The features are derived from both the Gabor cosine and sine filters. Through experiments, we demonstrate the effectiveness of a NN based classifier using Gabor features for identifying textures in a controlled environment. The neural network used for texture identification is based on the multilayer perceptr...
Neural-network-based fuzzy logic decision systems
Kulkarni, Arun D.; Giridhar, G. B.; Coca, Praveen
1994-10-01
During the last few years there has been a large and energetic upswing in research efforts aimed at synthesizing fuzzy logic with neural networks. This combination of neural networks and fuzzy logic seems natural because the two approaches generally attack the design of `intelligent' system from quite different angles. Neural networks provide algorithms for learning, classification, and optimization whereas fuzzy logic often deals with issues such as reasoning in a high (semantic or linguistic) level. Consequently the two technologies complement each other. In this paper, we combine neural networks with fuzzy logic techniques. We propose an artificial neural network (ANN) model for a fuzzy logic decision system. The model consists of six layers. The first three layers map the input variables to fuzzy set membership functions. The last three layers implement the decision rules. The model learns the decision rules using a supervised gradient descent procedure. As an illustration we considered two examples. The first example deals with pixel classification in multispectral satellite images. In our second example we used the fuzzy decision system to analyze data from magnetic resonance imaging (MRI) scans for tissue classification.
Energy Technology Data Exchange (ETDEWEB)
Mjalli, F.S.; Al-Asheh, S. [Chemical Engineering Department, Qatar University, Doha (Qatar)
2005-10-01
In this work advanced nonlinear neural networks based control system design algorithms are adopted to control a mechanistic model for an ethanol fermentation process. The process model equations for such systems are highly nonlinear. A neural network strategy has been implemented in this work for capturing the dynamics of the mechanistic model for the fermentation process. The neural network achieved has been validated against the mechanistic model. Two neural network based nonlinear control strategies have also been adopted using the model identified. The performance of the feedback linearization technique was compared to neural network model predictive control in terms of stability and set point tracking capabilities. Under servo conditions, the feedback linearization algorithm gave comparable tracking and stability. The feedback linearization controller achieved the control target faster than the model predictive one but with vigorous and sudden controller moves. (Abstract Copyright [2005], Wiley Periodicals, Inc.)
A NEURAL NETWORK BASED IRIS RECOGNITION SYSTEM FOR PERSONAL IDENTIFICATION
Directory of Open Access Journals (Sweden)
Usham Dias
2010-10-01
Full Text Available This paper presents biometric personal identification based on iris recognition using artificial neural networks. Personal identification system consists of localization of the iris region, normalization, enhancement and then iris pattern recognition using neural network. In this paper, through results obtained, we have shown that a person’s left and right eye are unique. In this paper, we also show that the network is sensitive to the initial weights and that over-training gives bad results. We also propose a fast algorithm for the localization of the inner and outer boundaries of the iris region. Results of simulations illustrate the effectiveness of the neural system in personal identification. Finally a hardware iris recognition model is proposed and implementation aspects are discussed.
Feature extraction for deep neural networks based on decision boundaries
Woo, Seongyoun; Lee, Chulhee
2017-05-01
Feature extraction is a process used to reduce data dimensions using various transforms while preserving the discriminant characteristics of the original data. Feature extraction has been an important issue in pattern recognition since it can reduce the computational complexity and provide a simplified classifier. In particular, linear feature extraction has been widely used. This method applies a linear transform to the original data to reduce the data dimensions. The decision boundary feature extraction method (DBFE) retains only informative directions for discriminating among the classes. DBFE has been applied to various parametric and non-parametric classifiers, which include the Gaussian maximum likelihood classifier (GML), the k-nearest neighbor classifier, support vector machines (SVM) and neural networks. In this paper, we apply DBFE to deep neural networks. This algorithm is based on the nonparametric version of DBFE, which was developed for neural networks. Experimental results with the UCI database show improved classification accuracy with reduced dimensionality.
Implementation of neural network based non-linear predictive
DEFF Research Database (Denmark)
Sørensen, Paul Haase; Nørgård, Peter Magnus; Ravn, Ole
1998-01-01
The paper describes a control method for non-linear systems based on generalized predictive control. Generalized predictive control (GPC) was developed to control linear systems including open loop unstable and non-minimum phase systems, but has also been proposed extended for the control of non......-linear systems. GPC is model-based and in this paper we propose the use of a neural network for the modeling of the system. Based on the neural network model a controller with extended control horizon is developed and the implementation issues are discussed, with particular emphasis on an efficient Quasi......-Newton optimization algorithm. The performance is demonstrated on a pneumatic servo system....
Spiking neural network-based control chart pattern recognition
Directory of Open Access Journals (Sweden)
Medhat H.A. Awadalla
2012-03-01
Full Text Available Due to an increasing competition in products, consumers have become more critical in choosing products. The quality of products has become more important. Statistical Process Control (SPC is usually used to improve the quality of products. Control charting plays the most important role in SPC. Control charts help to monitor the behavior of the process to determine whether it is stable or not. Unnatural patterns in control charts mean that there are some unnatural causes for variations in SPC. Spiking neural networks (SNNs are the third generation of artificial neural networks that consider time as an important feature for information representation and processing. In this paper, a spiking neural network architecture is proposed to be used for control charts pattern recognition (CCPR. Furthermore, enhancements to the SpikeProp learning algorithm are proposed. These enhancements provide additional learning rules for the synaptic delays, time constants and for the neurons thresholds. Simulated experiments have been conducted and the achieved results show a remarkable improvement in the overall performance compared with artificial neural networks.
RBF neural network based H∞ H∞ H∞ synchronization for ...
Indian Academy of Sciences (India)
Based on this neural network and linear matrix inequality (LMI) formulation, the RBFNNHS controller and the learning laws are presented to reduce the effect of disturbance to an H ∞ norm constraint. It is shown that ﬁnding the RBFNNHS controller and the learning laws can be transformed into the LMI problem and solved ...
RBF neural network based H∞ synchronization for unknown chaotic ...
Indian Academy of Sciences (India)
MS received 9 February 2010; accepted 24 May 2010. Abstract. In this paper, we propose a new H∞ synchronization strategy, called a. Radial Basis Function Neural Network H∞ synchronization (RBFNNHS) strategy, for unknown chaotic systems in the presence of external disturbance. In the pro- posed framework, a ...
An artificial neural network based fast radiative transfer model for ...
Indian Academy of Sciences (India)
In the present study, a fast radiative transfer model using neural networks is proposed to simulate radiances corresponding to the wavenumbers of INSAT-3D. Realistic atmospheric temperature and humidity profiles have been used for training the network. Spectral response functions of GOES-13, a satellite similar in ...
An artificial neural network based fast radiative transfer model for ...
Indian Academy of Sciences (India)
the present study, a fast radiative transfer model using neural networks is proposed to simulate radiances corresponding to the wavenumbers of ... in construction, purpose and design and already in use are used. The fast RT model is able to ... porates measurements from various instruments in comparison with other ...
Neural-network-based voice-tracking algorithm
Baker, Mary; Stevens, Charise; Chaparro, Brennen; Paschall, Dwayne
2002-11-01
A voice-tracking algorithm was developed and tested for the purposes of electronically separating the voice signals of simultaneous talkers. Many individuals suffer from hearing disorders that often inhibit their ability to focus on a single speaker in a multiple speaker environment (the cocktail party effect). Digital hearing aid technology makes it possible to implement complex algorithms for speech processing in both the time and frequency domains. In this work, an average magnitude difference function (AMDF) was performed on mixed voice signals in order to determine the fundamental frequencies present in the signals. A time prediction neural network was trained to recognize normal human voice inflection patterns, including rising, falling, rising-falling, and falling-rising patterns. The neural network was designed to track the fundamental frequency of a single talker based on the training procedure. The output of the neural network can be used to design an active filter for speaker segregation. Tests were done using audio mixing of two to three speakers uttering short phrases. The AMDF function accurately identified the fundamental frequencies present in the signal. The neural network was tested using a single speaker uttering a short sentence. The network accurately tracked the fundamental frequency of the speaker.
Neural network based system for script identification in Indian ...
Indian Academy of Sciences (India)
R. Narasimhan (Krishtel eMaging) 1461 1996 Oct 15 13:05:22
environments. The system developed includes a feature extractor and a modular neural network. The feature extractor consists of two stages. In the first stage ... environments is script/language identification (Muthusamy et al 1994; Hochberg et al 1997). ... In order to take advantage of the learning and generalization abilities ...
Modulation of grasping force in prosthetic hands using neural network-based predictive control.
Pasluosta, Cristian F; Chiu, Alan W L
2015-01-01
This chapter describes the implementation of a neural network-based predictive control system for driving a prosthetic hand. Nonlinearities associated with the electromechanical aspects of prosthetic devices present great challenges for precise control of this type of device. Model-based controllers may overcome this issue. Moreover, given the complexity of these kinds of electromechanical systems, neural network-based modeling arises as a good fit for modeling the fingers' dynamics. The results of simulations mimicking potential situations encountered during activities of daily living demonstrate the feasibility of this technique.
Neural network based PWM AC chopper fed induction motor drive
Directory of Open Access Journals (Sweden)
Venkatesan Jamuna
2009-01-01
Full Text Available In this paper, a new Simulink model for a neural network controlled PWM AC chopper fed single phase induction motor is proposed. Closed loop speed control is achieved using a neural network controller. To maintain a constant fluid flow with a variation in pressure head, drives like fan and pump are operated with closed loop speed control. The need to improve the quality and reliability of the drive circuit has increased because of the growing demand for improving the performance of motor drives. With the increased availability of MOSFET's and IGBT's, PWM converters can be used efficiently in low and medium power applications. From the simulation studies, it is seen that the PWM AC chopper has a better harmonic spectrum and lesser copper loss than the Phase controlled AC chopper. It is observed that the drive system with the proposed model produces better dynamic performance, reduced overshoot and fast transient response. .
Community structure of complex networks based on continuous neural network
Dai, Ting-ting; Shan, Chang-ji; Dong, Yan-shou
2017-09-01
As a new subject, the research of complex networks has attracted the attention of researchers from different disciplines. Community structure is one of the key structures of complex networks, so it is a very important task to analyze the community structure of complex networks accurately. In this paper, we study the problem of extracting the community structure of complex networks, and propose a continuous neural network (CNN) algorithm. It is proved that for any given initial value, the continuous neural network algorithm converges to the eigenvector of the maximum eigenvalue of the network modularity matrix. Therefore, according to the stability of the evolution of the network symbol will be able to get two community structure.
Neural Network-Based Abstract Generation for Opinions and Arguments
Wang, Lu; Ling, Wang
2016-01-01
We study the problem of generating abstractive summaries for opinionated text. We propose an attention-based neural network model that is able to absorb information from multiple text units to construct informative, concise, and fluent summaries. An importance-based sampling method is designed to allow the encoder to integrate information from an important subset of input. Automatic evaluation indicates that our system outperforms state-of-the-art abstractive and extractive summarization syst...
Cheng, Longlong; Zhang, Guangju; Wan, Baikun; Hao, Linlin; Qi, Hongzhi; Ming, Dong
2009-01-01
Functional electrical stimulation (FES) has been widely used in the area of neural engineering. It utilizes electrical current to activate nerves innervating extremities affected by paralysis. An effective combination of a traditional PID controller and a neural network, being capable of nonlinear expression and adaptive learning property, supply a more reliable approach to construct FES controller that help the paraplegia complete the action they want. A FES system tuned by Radial Basis Function (RBF) Neural Network-based Proportional-Integral-Derivative (PID) model was designed to control the knee joint according to the desired trajectory through stimulation of lower limbs muscles in this paper. Experiment result shows that the FES system with RBF Neural Network-based PID model get a better performance when tracking the preset trajectory of knee angle comparing with the system adjusted by Ziegler- Nichols tuning PID model.
Artificial neural network based approach to EEG signal simulation.
Tomasevic, Nikola M; Neskovic, Aleksandar M; Neskovic, Natasa J
2012-06-01
In this paper a new approach to the electroencephalogram (EEG) signal simulation based on the artificial neural networks (ANN) is proposed. The aim was to simulate the spontaneous human EEG background activity based solely on the experimentally acquired EEG data. Therefore, an EEG measurement campaign was conducted on a healthy awake adult in order to obtain an adequate ANN training data set. As demonstration of the performance of the ANN based approach, comparisons were made against autoregressive moving average (ARMA) filtering based method. Comprehensive quantitative and qualitative statistical analysis showed clearly that the EEG process obtained by the proposed method was in satisfactory agreement with the one obtained by measurements.
Deep Neural Network Based Demand Side Short Term Load Forecasting
Directory of Open Access Journals (Sweden)
Seunghyoung Ryu
2016-12-01
Full Text Available In the smart grid, one of the most important research areas is load forecasting; it spans from traditional time series analyses to recent machine learning approaches and mostly focuses on forecasting aggregated electricity consumption. However, the importance of demand side energy management, including individual load forecasting, is becoming critical. In this paper, we propose deep neural network (DNN-based load forecasting models and apply them to a demand side empirical load database. DNNs are trained in two different ways: a pre-training restricted Boltzmann machine and using the rectified linear unit without pre-training. DNN forecasting models are trained by individual customer’s electricity consumption data and regional meteorological elements. To verify the performance of DNNs, forecasting results are compared with a shallow neural network (SNN, a double seasonal Holt–Winters (DSHW model and the autoregressive integrated moving average (ARIMA. The mean absolute percentage error (MAPE and relative root mean square error (RRMSE are used for verification. Our results show that DNNs exhibit accurate and robust predictions compared to other forecasting models, e.g., MAPE and RRMSE are reduced by up to 17% and 22% compared to SNN and 9% and 29% compared to DSHW.
Artificial neural network based particle size prediction of polymeric nanoparticles.
Youshia, John; Ali, Mohamed Ehab; Lamprecht, Alf
2017-10-01
Particle size of nanoparticles and the respective polydispersity are key factors influencing their biopharmaceutical behavior in a large variety of therapeutic applications. Predicting these attributes would skip many preliminary studies usually required to optimize formulations. The aim was to build a mathematical model capable of predicting the particle size of polymeric nanoparticles produced by a pharmaceutical polymer of choice. Polymer properties controlling the particle size were identified as molecular weight, hydrophobicity and surface activity, and were quantified by measuring polymer viscosity, contact angle and interfacial tension, respectively. A model was built using artificial neural network including these properties as input with particle size and polydispersity index as output. The established model successfully predicted particle size of nanoparticles covering a range of 70-400nm prepared from other polymers. The percentage bias for particle prediction was 2%, 4% and 6%, for the training, validation and testing data, respectively. Polymer surface activity was found to have the highest impact on the particle size followed by viscosity and finally hydrophobicity. Results of this study successfully highlighted polymer properties affecting particle size and confirmed the usefulness of artificial neural networks in predicting the particle size and polydispersity of polymeric nanoparticles. Copyright © 2017 Elsevier B.V. All rights reserved.
Neural Network Based Model of an Industrial Oil-Fired Boiler System ...
African Journals Online (AJOL)
In this study, an oil-fired boiler system is modeled as a multivariable plant with two inputs (feed water rate and oil-fired flow rate) and two outputs (steam temperature and pressure). The plant parameters are modeled using artificial neural network, based on experimental data collected directly from the physical plant.
Stability Analysis of Neural Networks-Based System Identification
Directory of Open Access Journals (Sweden)
Talel Korkobi
2008-01-01
Full Text Available This paper treats some problems related to nonlinear systems identification. A stability analysis neural network model for identifying nonlinear dynamic systems is presented. A constrained adaptive stable backpropagation updating law is presented and used in the proposed identification approach. The proposed backpropagation training algorithm is modified to obtain an adaptive learning rate guarantying convergence stability. The proposed learning rule is the backpropagation algorithm under the condition that the learning rate belongs to a specified range defining the stability domain. Satisfying such condition, unstable phenomena during the learning process are avoided. A Lyapunov analysis leads to the computation of the expression of a convenient adaptive learning rate verifying the convergence stability criteria. Finally, the elaborated training algorithm is applied in several simulations. The results confirm the effectiveness of the CSBP algorithm.
Quantum neural network based machine translator for Hindi to English.
Narayan, Ravi; Singh, V P; Chakraverty, S
2014-01-01
This paper presents the machine learning based machine translation system for Hindi to English, which learns the semantically correct corpus. The quantum neural based pattern recognizer is used to recognize and learn the pattern of corpus, using the information of part of speech of individual word in the corpus, like a human. The system performs the machine translation using its knowledge gained during the learning by inputting the pair of sentences of Devnagri-Hindi and English. To analyze the effectiveness of the proposed approach, 2600 sentences have been evaluated during simulation and evaluation. The accuracy achieved on BLEU score is 0.7502, on NIST score is 6.5773, on ROUGE-L score is 0.9233, and on METEOR score is 0.5456, which is significantly higher in comparison with Google Translation and Bing Translation for Hindi to English Machine Translation.
Forecasting stochastic neural network based on financial empirical mode decomposition.
Wang, Jie; Wang, Jun
2017-06-01
In an attempt to improve the forecasting accuracy of stock price fluctuations, a new one-step-ahead model is developed in this paper which combines empirical mode decomposition (EMD) with stochastic time strength neural network (STNN). The EMD is a processing technique introduced to extract all the oscillatory modes embedded in a series, and the STNN model is established for considering the weight of occurrence time of the historical data. The linear regression performs the predictive availability of the proposed model, and the effectiveness of EMD-STNN is revealed clearly through comparing the predicted results with the traditional models. Moreover, a new evaluated method (q-order multiscale complexity invariant distance) is applied to measure the predicted results of real stock index series, and the empirical results show that the proposed model indeed displays a good performance in forecasting stock market fluctuations. Copyright © 2017 Elsevier Ltd. All rights reserved.
Directory of Open Access Journals (Sweden)
G. Vinodhini
2016-01-01
Full Text Available The aim of sentiment classification is to efficiently identify the emotions expressed in the form of text messages. Machine learning methods for sentiment classification have been extensively studied, due to their predominant classification performance. Recent studies suggest that ensemble based machine learning methods provide better performance in classification. Artificial neural networks (ANNs are rarely being investigated in the literature of sentiment classification. This paper compares neural network based sentiment classification methods (back propagation neural network (BPN, probabilistic neural network (PNN & homogeneous ensemble of PNN (HEN using varying levels of word granularity as features for feature level sentiment classification. They are validated using a dataset of product reviews collected from the Amazon reviews website. An empirical analysis is done to compare results of ANN based methods with two statistical individual methods. The methods are evaluated using five different quality measures and results show that the homogeneous ensemble of the neural network method provides better performance. Among the two neural network approaches used, probabilistic neural networks (PNNs outperform in classifying the sentiment of the product reviews. The integration of neural network based sentiment classification methods with principal component analysis (PCA as a feature reduction technique provides superior performance in terms of training time also.
Artificial Neural Network-Based System for PET Volume Segmentation
Directory of Open Access Journals (Sweden)
Mhd Saeed Sharif
2010-01-01
Full Text Available Tumour detection, classification, and quantification in positron emission tomography (PET imaging at early stage of disease are important issues for clinical diagnosis, assessment of response to treatment, and radiotherapy planning. Many techniques have been proposed for segmenting medical imaging data; however, some of the approaches have poor performance, large inaccuracy, and require substantial computation time for analysing large medical volumes. Artificial intelligence (AI approaches can provide improved accuracy and save decent amount of time. Artificial neural networks (ANNs, as one of the best AI techniques, have the capability to classify and quantify precisely lesions and model the clinical evaluation for a specific problem. This paper presents a novel application of ANNs in the wavelet domain for PET volume segmentation. ANN performance evaluation using different training algorithms in both spatial and wavelet domains with a different number of neurons in the hidden layer is also presented. The best number of neurons in the hidden layer is determined according to the experimental results, which is also stated Levenberg-Marquardt backpropagation training algorithm as the best training approach for the proposed application. The proposed intelligent system results are compared with those obtained using conventional techniques including thresholding and clustering based approaches. Experimental and Monte Carlo simulated PET phantom data sets and clinical PET volumes of nonsmall cell lung cancer patients were utilised to validate the proposed algorithm which has demonstrated promising results.
A neural network based reputation bootstrapping approach for service selection
Wu, Quanwang; Zhu, Qingsheng; Li, Peng
2015-10-01
With the concept of service-oriented computing becoming widely accepted in enterprise application integration, more and more computing resources are encapsulated as services and published online. Reputation mechanism has been studied to establish trust on prior unknown services. One of the limitations of current reputation mechanisms is that they cannot assess the reputation of newly deployed services as no record of their previous behaviours exists. Most of the current bootstrapping approaches merely assign default reputation values to newcomers. However, by this kind of methods, either newcomers or existing services will be favoured. In this paper, we present a novel reputation bootstrapping approach, where correlations between features and performance of existing services are learned through an artificial neural network (ANN) and they are then generalised to establish a tentative reputation when evaluating new and unknown services. Reputations of services published previously by the same provider are also incorporated for reputation bootstrapping if available. The proposed reputation bootstrapping approach is seamlessly embedded into an existing reputation model and implemented in the extended service-oriented architecture. Empirical studies of the proposed approach are shown at last.
Neural network based optimal control of HVAC&R systems
Ning, Min
Heating, Ventilation, Air-Conditioning and Refrigeration (HVAC&R) systems have wide applications in providing a desired indoor environment for different types of buildings. It is well acknowledged that 30%-40% of the total energy generated is consumed by buildings and HVAC&R systems alone account for more than 50% of the building energy consumption. Low operational efficiency especially under partial load conditions and poor control are part of reasons for such high energy consumption. To improve energy efficiency, HVAC&R systems should be properly operated to maintain a comfortable and healthy indoor environment under dynamic ambient and indoor conditions with the least energy consumption. This research focuses on the optimal operation of HVAC&R systems. The optimization problem is formulated and solved to find the optimal set points for the chilled water supply temperature, discharge air temperature and AHU (air handling unit) fan static pressure such that the indoor environment is maintained with the least chiller and fan energy consumption. To achieve this objective, a dynamic system model is developed first to simulate the system behavior under different control schemes and operating conditions. The system model is modular in structure, which includes a water-cooled vapor compression chiller model and a two-zone VAV system model. A fuzzy-set based extended transformation approach is then applied to investigate the uncertainties of this model caused by uncertain parameters and the sensitivities of the control inputs with respect to the interested model outputs. A multi-layer feed forward neural network is constructed and trained in unsupervised mode to minimize the cost function which is comprised of overall energy cost and penalty cost when one or more constraints are violated. After training, the network is implemented as a supervisory controller to compute the optimal settings for the system. In order to implement the optimal set points predicted by the
A Neural Network Based Hybrid Mixture Model to Extract Information from Non-linear Mixed Pixels
Directory of Open Access Journals (Sweden)
Uttam Kumar
2012-09-01
Full Text Available Signals acquired by sensors in the real world are non-linear combinations, requiring non-linear mixture models to describe the resultant mixture spectra for the endmember’s (pure pixel’s distribution. This communication discusses inferring class fraction through a novel hybrid mixture model (HMM. HMM is a three-step process, where the endmembers are first derived from the images themselves using the N-FINDR algorithm. These endmembers are used by the linear mixture model (LMM in the second step that provides an abundance estimation in a linear fashion. Finally, the abundance values along with the training samples representing the actual ground proportions are fed into neural network based multi-layer perceptron (MLP architecture as input to train the neurons. The neural output further refines the abundance estimates to account for the non-linear nature of the mixing classes of interest. HMM is first implemented and validated on simulated hyper spectral data of 200 bands and subsequently on real time MODIS data with a spatial resolution of 250 m. The results on computer simulated data show that the method gives acceptable results for unmixing pixels with an overall RMSE of 0.0089 ± 0.0022 with LMM and 0.0030 ± 0.0001 with the HMM when compared to actual class proportions. The unmixed MODIS images showed overall RMSE with HMM as 0.0191 ± 0.022 as compared to the LMM output considered alone that had an overall RMSE of 0.2005 ± 0.41, indicating that individual class abundances obtained from HMM are very close to the real observations.
Sun, Ran; Wang, Jihe; Zhang, Dexin; Shao, Xiaowei
2018-02-01
This paper presents an adaptive neural networks-based control method for spacecraft formation with coupled translational and rotational dynamics using only aerodynamic forces. It is assumed that each spacecraft is equipped with several large flat plates. A coupled orbit-attitude dynamic model is considered based on the specific configuration of atmospheric-based actuators. For this model, a neural network-based adaptive sliding mode controller is implemented, accounting for system uncertainties and external perturbations. To avoid invalidation of the neural networks destroying stability of the system, a switching control strategy is proposed which combines an adaptive neural networks controller dominating in its active region and an adaptive sliding mode controller outside the neural active region. An optimal process is developed to determine the control commands for the plates system. The stability of the closed-loop system is proved by a Lyapunov-based method. Comparative results through numerical simulations illustrate the effectiveness of executing attitude control while maintaining the relative motion, and higher control accuracy can be achieved by using the proposed neural-based switching control scheme than using only adaptive sliding mode controller.
An Efficient Neural Network Based Modeling Method for Automotive EMC Simulation
Frank, Florian; Weigel, Robert
2011-09-01
This paper presents a newly developed methodology for VHDL-AMS model integration into SPICE-based EMC simulations. To this end the VHDL-AMS model, which is available in a compiled version only, is characterized under typical loading conditions, and afterwards a neural network based technique is applied to convert characteristic voltage and current data into an equivalent circuit in SPICE syntax. After the explanation of the whole method and the presentation of a newly developed switched state space dynamic neural network model, the entire analysis process is demonstrated using a typical application from automotive industry.
Paul, R R; Mukherjee, A; Dutta, P K; Banerjee, S; Pal, M; Chatterjee, J; Chaudhuri, K; Mukkerjee, K
2005-01-01
Aim: To describe a novel neural network based oral precancer (oral submucous fibrosis; OSF) stage detection method. Method: The wavelet coefficients of transmission electron microscopy images of collagen fibres from normal oral submucosa and OSF tissues were used to choose the feature vector which, in turn, was used to train the artificial neural network. Results: The trained network was able to classify normal and oral precancer stages (less advanced and advanced) after obtaining the image as an input. Conclusions: The results obtained from this proposed technique were promising and suggest that with further optimisation this method could be used to detect and stage OSF, and could be adapted for other conditions. PMID:16126873
Grantham, Katie
2003-01-01
Reusable Launch Vehicles (RLVs) have different mission requirements than the Space Shuttle, which is used for benchmark guidance design. Therefore, alternative Terminal Area Energy Management (TAEM) and Approach and Landing (A/L) Guidance schemes can be examined in the interest of cost reduction. A neural network based solution for a finite horizon trajectory optimization problem is presented in this paper. In this approach the optimal trajectory of the vehicle is produced by adaptive critic based neural networks, which were trained off-line to maintain a gradual glideslope.
A Predictive Neural Network-Based Cascade Control for pH Reactors
Directory of Open Access Journals (Sweden)
Mujahed AlDhaifallah
2016-01-01
Full Text Available This paper is concerned with the development of predictive neural network-based cascade control for pH reactors. The cascade structure consists of a master control loop (fuzzy proportional-integral and a slave one (predictive neural network. The master loop is chosen to be more accurate but slower than the slave one. The strong features found in cascade structure have been added to the inherent features in model predictive neural network. The neural network is used to alleviate modeling difficulties found with pH reactor and to predict its behavior. The parameters of predictive algorithm are determined using an optimization algorithm. The effectiveness and feasibility of the proposed design have been demonstrated using MatLab.
Directory of Open Access Journals (Sweden)
Eline Janssens
2016-11-01
Full Text Available Speed is an important parameter of an inspection system. Inline computed tomography systems exist but are generally expensive. Moreover, their throughput is limited by the speed of the reconstruction algorithm. In this work, we propose a Neural Network-based Hilbert transform Filtered Backprojection (NN-hFBP method to reconstruct objects in an inline scanning environment in a fast and accurate way. Experiments based on apple X-ray scans show that the NN-hFBP method allows to reconstruct images with a substantially better tradeoff between image quality and reconstruction time.
Extended neural network-based scheme for real-time force tracking with magnetorheological dampers
DEFF Research Database (Denmark)
Weber, Felix; Bhowmik, Subrata; Høgsberg, Jan Becker
2014-01-01
This paper validates numerically and experimentally a new neural network-based real-time force tracking scheme for magnetorheological (MR) dampers on a five-storey shear frame with MR damper. The inverse model is trained with absolute values of measured velocity and force because the targeted...... current is a positive quantity. The validation shows accurate results except of small current spikes when the desired force is in the vicinity of the residual MR damper force. In the closed-loop, higher frequency components in the current are triggered by the transition of the actual MR damper force from...
An Artificial Neural Network Based Robot Controller that Uses Rat’s Brain Signals
Directory of Open Access Journals (Sweden)
Marsel Mano
2013-04-01
Full Text Available Brain machine interface (BMI has been proposed as a novel technique to control prosthetic devices aimed at restoring motor functions in paralyzed patients. In this paper, we propose a neural network based controller that maps rat’s brain signals and transforms them into robot movement. First, the rat is trained to move the robot by pressing the right and left lever in order to get food. Next, we collect brain signals with four implanted electrodes, two in the motor cortex and two in the somatosensory cortex area. The collected data are used to train and evaluate different artificial neural controllers. Trained neural controllers are employed online to map brain signals and transform them into robot motion. Offline and online classification results of rat’s brain signals show that the Radial Basis Function Neural Networks (RBFNN outperforms other neural networks. In addition, online robot control results show that even with a limited number of electrodes, the robot motion generated by RBFNN matched the motion generated by the left and right lever position.
Validity-Guided Fuzzy Clustering Evaluation for Neural Network-Based Time-Frequency Reassignment
Directory of Open Access Journals (Sweden)
Ahmad Khan Adnan
2010-01-01
Full Text Available Abstract This paper describes the validity-guided fuzzy clustering evaluation for optimal training of localized neural networks (LNNs used for reassigning time-frequency representations (TFRs. Our experiments show that the validity-guided fuzzy approach ameliorates the difficulty of choosing correct number of clusters and in conjunction with neural network-based processing technique utilizing a hybrid approach can effectively reduce the blur in the spectrograms. In the course of every partitioning problem the number of subsets must be given before the calculation, but it is rarely known apriori, in this case it must be searched also with using validity measures. Experimental results demonstrate the effectiveness of the approach.
Singh, H P; Sukavanam, N
2012-01-01
This paper proposes a new adaptive neural network based control scheme for switched linear systems with parametric uncertainty and external disturbance. A key feature of this scheme is that the prior information of the possible upper bound of the uncertainty is not required. A feedforward neural network is employed to learn this upper bound. The adaptive learning algorithm is derived from Lyapunov stability analysis so that the system response under arbitrary switching laws is guaranteed uniformly ultimately bounded. A comparative simulation study with robust controller given in [Zhang L, Lu Y, Chen Y, Mastorakis NE. Robust uniformly ultimate boundedness control for uncertain switched linear systems. Computers and Mathematics with Applications 2008; 56: 1709-14] is presented. Copyright © 2011 ISA. Published by Elsevier Ltd. All rights reserved.
DEFF Research Database (Denmark)
Bhowmik, Subrata
2011-01-01
This paper presents a neural network based semi-active control method for a rotary type magnetorheological (MR) damper. The characteristics of the MR damper are described by the classic Bouc-Wen model, and the performance of the proposed control method is evaluated in terms of a base exited shear...... frame structure. As demonstrated in the literature effective damping of flexible structures is obtained by a suitable combination of pure friction and negative damper stiffness. This damper model is rate-independent and fully described by the desired shape of the hysteresis loops or force...... mode of the structure. The neural network control is then developed to reproduce the desired force based on damper displacement and velocity as network input, and it is therefore referred to as an amplitude dependent model reference control method. An inverse model of the MR damper is needed...
Decoherence and Entanglement Simulation in a Model of Quantum Neural Network Based on Quantum Dots
Directory of Open Access Journals (Sweden)
Altaisky Mikhail V.
2016-01-01
Full Text Available We present the results of the simulation of a quantum neural network based on quantum dots using numerical method of path integral calculation. In the proposed implementation of the quantum neural network using an array of single-electron quantum dots with dipole-dipole interaction, the coherence is shown to survive up to 0.1 nanosecond in time and up to the liquid nitrogen temperature of 77K.We study the quantum correlations between the quantum dots by means of calculation of the entanglement of formation in a pair of quantum dots on the GaAs based substrate with dot size of 100 ÷ 101 nanometer and interdot distance of 101 ÷ 102 nanometers order.
Directory of Open Access Journals (Sweden)
Manjunath Patel Gowdru Chandrashekarappa
2014-01-01
Full Text Available The present research work is focussed to develop an intelligent system to establish the input-output relationship utilizing forward and reverse mappings of artificial neural networks. Forward mapping aims at predicting the density and secondary dendrite arm spacing (SDAS from the known set of squeeze cast process parameters such as time delay, pressure duration, squeezes pressure, pouring temperature, and die temperature. An attempt is also made to meet the industrial requirements of developing the reverse model to predict the recommended squeeze cast parameters for the desired density and SDAS. Two different neural network based approaches have been proposed to carry out the said task, namely, back propagation neural network (BPNN and genetic algorithm neural network (GA-NN. The batch mode of training is employed for both supervised learning networks and requires huge training data. The requirement of huge training data is generated artificially at random using regression equation derived through real experiments carried out earlier by the same authors. The performances of BPNN and GA-NN models are compared among themselves with those of regression for ten test cases. The results show that both models are capable of making better predictions and the models can be effectively used in shop floor in selection of most influential parameters for the desired outputs.
Nayar, Priya; Singh, Bhim; Mishra, Sukumar
2017-08-01
An artificial intelligence based control algorithm is used in solving power quality problems of a diesel engine driven synchronous generator with automatic voltage regulator and governor based standalone system. A voltage source converter integrated with a battery energy storage system is employed to mitigate the power quality problems. An adaptive neural network based signed regressor control algorithm is used for the estimation of the fundamental component of load currents for control of a standalone system with load leveling as an integral feature. The developed model of the system performs accurately under varying load conditions and provides good dynamic response to the step changes in loads. The real time performance is achieved using MATLAB along with simulink/simpower system toolboxes and results adhere to an IEEE-519 standard for power quality enhancement.
Kim, Nakwan
Utilizing the universal approximation property of neural networks, we develop several novel approaches to neural network-based adaptive output feedback control of nonlinear systems, and illustrate these approaches for several flight control applications. In particular, we address the problem of non-affine systems and eliminate the fixed point assumption present in earlier work. All of the stability proofs are carried out in a form that eliminates an algebraic loop in the neural network implementation. An approximate input/output feedback linearizing controller is augmented with a neural network using input/output sequences of the uncertain system. These approaches permit adaptation to both parametric uncertainty and unmodeled dynamics. All physical systems also have control position and rate limits, which may either deteriorate performance or cause instability for a sufficiently high control bandwidth. Here we apply a method for protecting an adaptive process from the effects of input saturation and time delays, known as "pseudo control hedging". This method was originally developed for the state feedback case, and we provide a stability analysis that extends its domain of applicability to the case of output feedback. The approach is illustrated by the design of a pitch-attitude flight control system for a linearized model of an R-50 experimental helicopter, and by the design of a pitch-rate control system for a 58-state model of a flexible aircraft consisting of rigid body dynamics coupled with actuator and flexible modes. A new approach to augmentation of an existing linear controller is introduced. It is especially useful when there is limited information concerning the plant model, and the existing controller. The approach is applied to the design of an adaptive autopilot for a guided munition. Design of a neural network adaptive control that ensures asymptotically stable tracking performance is also addressed.
Lee, Jong Jae; Lee, Jong Won; Yi, Jin Hak; Yun, Chung Bang; Jung, Hie Young
2005-02-01
Structural health monitoring has become an important research topic in conjunction with damage assessment and safety evaluation of structures. The use of system identification approaches for damage detection has been expanded in recent years owing to the advancements in signal analysis and information processing techniques. Soft computing techniques such as neural networks and genetic algorithm have been utilized increasingly for this end due to their excellent pattern recognition capability. In this study, a neural networks-based damage detection method using the modal properties is presented, which can effectively consider the modelling errors in the baseline finite element model from which the training patterns are to be generated. The differences or the ratios of the mode shape components between before and after damage are used as the input to the neural networks in this method, since they are found to be less sensitive to the modelling errors than the mode shapes themselves. Two numerical example analyses on a simple beam and a multi-girder bridge are presented to demonstrate the effectiveness of the proposed method. Results of laboratory test on a simply supported bridge model and field test on a bridge with multiple girders confirm the applicability of the present method.
Neural-Network-Based Fuzzy Logic Navigation Control for Intelligent Vehicles
Directory of Open Access Journals (Sweden)
Ahcene Farah
2002-06-01
Full Text Available This paper proposes a Neural-Network-Based Fuzzy logic system for navigation control of intelligent vehicles. First, the use of Neural Networks and Fuzzy Logic to provide intelligent vehicles with more autonomy and intelligence is discussed. Second, the system for the obstacle avoidance behavior is developed. Fuzzy Logic improves Neural Networks (NN obstacle avoidance approach by handling imprecision and rule-based approximate reasoning. This system must make the vehicle able, after supervised learning, to achieve two tasks: 1- to make one’s way towards its target by a NN, and 2- to avoid static or dynamic obstacles by a Fuzzy NN capturing the behavior of a human expert. Afterwards, two association phases between each task and the appropriate actions are carried out by Trial and Error learning and their coordination allows to decide the appropriate action. Finally, the simulation results display the generalization and adaptation abilities of the system by testing it in new unexplored environments.
A robust neural network-based approach for microseismic event detection
Akram, Jubran
2017-08-17
We present an artificial neural network based approach for robust event detection from low S/N waveforms. We use a feed-forward network with a single hidden layer that is tuned on a training dataset and later applied on the entire example dataset for event detection. The input features used include the average of absolute amplitudes, variance, energy-ratio and polarization rectilinearity. These features are calculated in a moving-window of same length for the entire waveform. The output is set as a user-specified relative probability curve, which provides a robust way of distinguishing between weak and strong events. An optimal network is selected by studying the weight-based saliency and effect of number of neurons on the predicted results. Using synthetic data examples, we demonstrate that this approach is effective in detecting weaker events and reduces the number of false positives.
Yang, Xinyu; Xu, Guoai; Li, Qi; Guo, Yanhui; Zhang, Miao
2017-01-01
Authorship attribution is to identify the most likely author of a given sample among a set of candidate known authors. It can be not only applied to discover the original author of plain text, such as novels, blogs, emails, posts etc., but also used to identify source code programmers. Authorship attribution of source code is required in diverse applications, ranging from malicious code tracking to solving authorship dispute or software plagiarism detection. This paper aims to propose a new method to identify the programmer of Java source code samples with a higher accuracy. To this end, it first introduces back propagation (BP) neural network based on particle swarm optimization (PSO) into authorship attribution of source code. It begins by computing a set of defined feature metrics, including lexical and layout metrics, structure and syntax metrics, totally 19 dimensions. Then these metrics are input to neural network for supervised learning, the weights of which are output by PSO and BP hybrid algorithm. The effectiveness of the proposed method is evaluated on a collected dataset with 3,022 Java files belong to 40 authors. Experiment results show that the proposed method achieves 91.060% accuracy. And a comparison with previous work on authorship attribution of source code for Java language illustrates that this proposed method outperforms others overall, also with an acceptable overhead.
Deep Recurrent Neural Network-Based Autoencoders for Acoustic Novelty Detection
Directory of Open Access Journals (Sweden)
Erik Marchi
2017-01-01
Full Text Available In the emerging field of acoustic novelty detection, most research efforts are devoted to probabilistic approaches such as mixture models or state-space models. Only recent studies introduced (pseudo-generative models for acoustic novelty detection with recurrent neural networks in the form of an autoencoder. In these approaches, auditory spectral features of the next short term frame are predicted from the previous frames by means of Long-Short Term Memory recurrent denoising autoencoders. The reconstruction error between the input and the output of the autoencoder is used as activation signal to detect novel events. There is no evidence of studies focused on comparing previous efforts to automatically recognize novel events from audio signals and giving a broad and in depth evaluation of recurrent neural network-based autoencoders. The present contribution aims to consistently evaluate our recent novel approaches to fill this white spot in the literature and provide insight by extensive evaluations carried out on three databases: A3Novelty, PASCAL CHiME, and PROMETHEUS. Besides providing an extensive analysis of novel and state-of-the-art methods, the article shows how RNN-based autoencoders outperform statistical approaches up to an absolute improvement of 16.4% average F-measure over the three databases.
Geramipour, A; Makki, S; Erfanian, A
2015-01-01
Individuals with spinal cord injury or neurological disorders have problems in urinary bladder storage and in voiding function. In these people, the detrusor of bladder contracts at low volume and this causes incontinence. The goal of bladder control is to increase the bladder capacity by electrical stimulation of relative nerves such as pelvic nerves, sacral nerve roots or pudendal nerves. For this purpose, the bladder pressure has to be monitored continuously. In this paper, we propose a method for real-time estimating the bladder pressure using artificial neural network. The method is based upon measurements of electroneurogram (ENG) signal of pudendal nerve. This approach yields synthetic bladder pressure estimates during bladder contraction. The experiments were conducted on three rats. The results show that neural predictor can provide accurate estimation and prediction of bladder pressure with good generalization ability. The average error of 1-second and 5-second ahead prediction of bladder pressure are 9.62% and 10.54%, respectively.
Barzegar, Rahim; Fijani, Elham; Asghari Moghaddam, Asghar; Tziritis, Evangelos
2017-12-01
Accurate prediction of groundwater level (GWL) fluctuations can play an important role in water resources management. The aims of the research are to evaluate the performance of different hybrid wavelet-group method of data handling (WA-GMDH) and wavelet-extreme learning machine (WA-ELM) models and to combine different wavelet based models for forecasting the GWL for one, two and three months step-ahead in the Maragheh-Bonab plain, NW Iran, as a case study. The research used totally 367 monthly GWLs (m) datasets (Sep 1985-Mar 2016) which were split into two subsets; the first 312 datasets (85% of total) were used for model development (training) and the remaining 55 ones (15% of total) for model evaluation (testing). The stepwise selection was used to select appropriate lag times as the inputs of the proposed models. The performance criteria such as coefficient of determination (R2), root mean square error (RMSE) and Nash-Sutcliffe efficiency coefficient (NSC) were used for assessing the efficiency of the models. The results indicated that the ELM models outperformed GMDH models. To construct the hybrid wavelet based models, the inputs and outputs were decomposed into sub-time series employing different maximal overlap discrete wavelet transform (MODWT) functions, namely Daubechies, Symlet, Haar and Dmeyer of different orders at level two. Subsequently, these sub-time series were served in the GMDH and ELM models as an input dataset to forecast the multi-step-ahead GWL. The wavelet based models improved the performances of GMDH and ELM models for multi-step-ahead GWL forecasting. To combine the advantages of different wavelets, a least squares boosting (LSBoost) algorithm was applied. The use of the boosting multi-WA-neural network models provided the best performances for GWL forecasts in comparison with single WA-neural network-based models. Copyright © 2017 Elsevier B.V. All rights reserved.
A neural network-based exploratory learning and motor planning system for co-robots
Directory of Open Access Journals (Sweden)
Byron V Galbraith
2015-07-01
Full Text Available Collaborative robots, or co-robots, are semi-autonomous robotic agents designed to work alongside humans in shared workspaces. To be effective, co-robots require the ability to respond and adapt to dynamic scenarios encountered in natural environments. One way to achieve this is through exploratory learning, or learning by doing, an unsupervised method in which co-robots are able to build an internal model for motor planning and coordination based on real-time sensory inputs. In this paper, we present an adaptive neural network-based system for co-robot control that employs exploratory learning to achieve the coordinated motor planning needed to navigate toward, reach for, and grasp distant objects. To validate this system we used the 11-degrees-of-freedom RoPro Calliope mobile robot. Through motor babbling of its wheels and arm, the Calliope learned how to relate visual and proprioceptive information to achieve hand-eye-body coordination. By continually evaluating sensory inputs and externally provided goal directives, the Calliope was then able to autonomously select the appropriate wheel and joint velocities needed to perform its assigned task, such as following a moving target or retrieving an indicated object.
A neural network-based exploratory learning and motor planning system for co-robots.
Galbraith, Byron V; Guenther, Frank H; Versace, Massimiliano
2015-01-01
Collaborative robots, or co-robots, are semi-autonomous robotic agents designed to work alongside humans in shared workspaces. To be effective, co-robots require the ability to respond and adapt to dynamic scenarios encountered in natural environments. One way to achieve this is through exploratory learning, or "learning by doing," an unsupervised method in which co-robots are able to build an internal model for motor planning and coordination based on real-time sensory inputs. In this paper, we present an adaptive neural network-based system for co-robot control that employs exploratory learning to achieve the coordinated motor planning needed to navigate toward, reach for, and grasp distant objects. To validate this system we used the 11-degrees-of-freedom RoPro Calliope mobile robot. Through motor babbling of its wheels and arm, the Calliope learned how to relate visual and proprioceptive information to achieve hand-eye-body coordination. By continually evaluating sensory inputs and externally provided goal directives, the Calliope was then able to autonomously select the appropriate wheel and joint velocities needed to perform its assigned task, such as following a moving target or retrieving an indicated object.
A novel neural network-based technique for smart gas sensors operating in a dynamic environment.
Baha, Hakim; Dibi, Zohir
2009-01-01
Thanks to their high sensitivity and low-cost, metal oxide gas sensors (MOX) are widely used in gas detection, although they present well-known problems (lack of selectivity and environmental effects…). We present in this paper a novel neural network- based technique to remedy these problems. The idea is to create intelligent models; the first one, called corrector, can automatically linearize a sensor's response characteristics and eliminate its dependency on the environmental parameters. The corrector's responses are processed with the second intelligent model which has the role of discriminating exactly the detected gas (nature and concentration). The gas sensors used are industrial resistive kind (TGS8xx, by Figaro Engineering). The MATLAB environment is used during the design phase and optimization. The sensor models, the corrector, and the selective model were implemented and tested in the PSPICE simulator. The sensor model accurately expresses the nonlinear character of the response and the dependence on temperature and relative humidity in addition to their gas nature dependency. The corrector linearizes and compensates the sensor's responses. The method discriminates qualitatively and quantitatively between seven gases. The advantage of the method is that it uses a small representative database so we can easily implement the model in an electrical simulator. This method can be extended to other sensors.
Acceleration of spiking neural network based pattern recognition on NVIDIA graphics processors.
Han, Bing; Taha, Tarek M
2010-04-01
There is currently a strong push in the research community to develop biological scale implementations of neuron based vision models. Systems at this scale are computationally demanding and generally utilize more accurate neuron models, such as the Izhikevich and the Hodgkin-Huxley models, in favor of the more popular integrate and fire model. We examine the feasibility of using graphics processing units (GPUs) to accelerate a spiking neural network based character recognition network to enable such large scale systems. Two versions of the network utilizing the Izhikevich and Hodgkin-Huxley models are implemented. Three NVIDIA general-purpose (GP) GPU platforms are examined, including the GeForce 9800 GX2, the Tesla C1060, and the Tesla S1070. Our results show that the GPGPUs can provide significant speedup over conventional processors. In particular, the fastest GPGPU utilized, the Tesla S1070, provided a speedup of 5.6 and 84.4 over highly optimized implementations on the fastest central processing unit (CPU) tested, a quadcore 2.67 GHz Xeon processor, for the Izhikevich and the Hodgkin-Huxley models, respectively. The CPU implementation utilized all four cores and the vector data parallelism offered by the processor. The results indicate that GPUs are well suited for this application domain.
Quan, Hao; Srinivasan, Dipti; Khosravi, Abbas
2015-09-01
Penetration of renewable energy resources, such as wind and solar power, into power systems significantly increases the uncertainties on system operation, stability, and reliability in smart grids. In this paper, the nonparametric neural network-based prediction intervals (PIs) are implemented for forecast uncertainty quantification. Instead of a single level PI, wind power forecast uncertainties are represented in a list of PIs. These PIs are then decomposed into quantiles of wind power. A new scenario generation method is proposed to handle wind power forecast uncertainties. For each hour, an empirical cumulative distribution function (ECDF) is fitted to these quantile points. The Monte Carlo simulation method is used to generate scenarios from the ECDF. Then the wind power scenarios are incorporated into a stochastic security-constrained unit commitment (SCUC) model. The heuristic genetic algorithm is utilized to solve the stochastic SCUC problem. Five deterministic and four stochastic case studies incorporated with interval forecasts of wind power are implemented. The results of these cases are presented and discussed together. Generation costs, and the scheduled and real-time economic dispatch reserves of different unit commitment strategies are compared. The experimental results show that the stochastic model is more robust than deterministic ones and, thus, decreases the risk in system operations of smart grids.
A Novel Neural Network-Based Technique for Smart Gas Sensors Operating in a Dynamic Environment
Directory of Open Access Journals (Sweden)
Zohir Dibi
2009-11-01
Full Text Available Thanks to their high sensitivity and low-cost, metal oxide gas sensors (MOX are widely used in gas detection, although they present well-known problems (lack of selectivity and environmental effects…. We present in this paper a novel neural network- based technique to remedy these problems. The idea is to create intelligent models; the first one, called corrector, can automatically linearize a sensor’s response characteristics and eliminate its dependency on the environmental parameters. The corrector’s responses are processed with the second intelligent model which has the role of discriminating exactly the detected gas (nature and concentration. The gas sensors used are industrial resistive kind (TGS8xx, by Figaro Engineering. The MATLAB environment is used during the design phase and optimization. The sensor models, the corrector, and the selective model were implemented and tested in the PSPICE simulator. The sensor model accurately expresses the nonlinear character of the response and the dependence on temperature and relative humidity in addition to their gas nature dependency. The corrector linearizes and compensates the sensor’s responses. The method discriminates qualitatively and quantitatively between seven gases. The advantage of the method is that it uses a small representative database so we can easily implement the model in an electrical simulator. This method can be extended to other sensors.
Neural Network Based on Quantum Chemistry for Predicting Melting Point of Organic Compounds
Lazzús, Juan A.
2009-02-01
The melting points of organic compounds were estimated using a combined method that includes a backpropagation neural network and quantitative structure property relationship (QSPR) parameters in quantum chemistry. Eleven descriptors that reflect the intermolecular forces and molecular symmetry were used as input variables. QSPR parameters were calculated using molecular modeling and PM3 semi-empirical molecular orbital theories. A total of 260 compounds were used to train the network, which was developed using MatLab. Then, the melting points of 73 other compounds were predicted and results were compared to experimental data from the literature. The study shows that the chosen artificial neural network and the quantitative structure property relationships method present an excellent alternative for the estimation of the melting point of an organic compound, with average absolute deviation of 5%.
Kang, Tianyu; Ding, Wei; Zhang, Luoyan; Ziemek, Daniel; Zarringhalam, Kourosh
2017-12-19
Stratification of patient subpopulations that respond favorably to treatment or experience and adverse reaction is an essential step toward development of new personalized therapies and diagnostics. It is currently feasible to generate omic-scale biological measurements for all patients in a study, providing an opportunity for machine learning models to identify molecular markers for disease diagnosis and progression. However, the high variability of genetic background in human populations hampers the reproducibility of omic-scale markers. In this paper, we develop a biological network-based regularized artificial neural network model for prediction of phenotype from transcriptomic measurements in clinical trials. To improve model sparsity and the overall reproducibility of the model, we incorporate regularization for simultaneous shrinkage of gene sets based on active upstream regulatory mechanisms into the model. We benchmark our method against various regression, support vector machines and artificial neural network models and demonstrate the ability of our method in predicting the clinical outcomes using clinical trial data on acute rejection in kidney transplantation and response to Infliximab in ulcerative colitis. We show that integration of prior biological knowledge into the classification as developed in this paper, significantly improves the robustness and generalizability of predictions to independent datasets. We provide a Java code of our algorithm along with a parsed version of the STRING DB database. In summary, we present a method for prediction of clinical phenotypes using baseline genome-wide expression data that makes use of prior biological knowledge on gene-regulatory interactions in order to increase robustness and reproducibility of omic-scale markers. The integrated group-wise regularization methods increases the interpretability of biological signatures and gives stable performance estimates across independent test sets.
Directory of Open Access Journals (Sweden)
Vincent A Emanuele
Full Text Available SELDI-TOF mass spectrometer's compact size and automated, high throughput design have been attractive to clinical researchers, and the platform has seen steady-use in biomarker studies. Despite new algorithms and preprocessing pipelines that have been developed to address reproducibility issues, visual inspection of the results of SELDI spectra preprocessing by the best algorithms still shows miscalled peaks and systematic sources of error. This suggests that there continues to be problems with SELDI preprocessing. In this work, we study the preprocessing of SELDI in detail and introduce improvements. While many algorithms, including the vendor supplied software, can identify peak clusters of specific mass (or m/z in groups of spectra with high specificity and low false discover rate (FDR, the algorithms tend to underperform estimating the exact prevalence and intensity of peaks in those clusters. Thus group differences that at first appear very strong are shown, after careful and laborious hand inspection of the spectra, to be less than significant. Here we introduce a wavelet/neural network based algorithm which mimics what a team of expert, human users would call for peaks in each of several hundred spectra in a typical SELDI clinical study. The wavelet denoising part of the algorithm optimally smoothes the signal in each spectrum according to an improved suite of signal processing algorithms previously reported (the LibSELDI toolbox under development. The neural network part of the algorithm combines those results with the raw signal and a training dataset of expertly called peaks, to call peaks in a test set of spectra with approximately 95% accuracy. The new method was applied to data collected from a study of cervical mucus for the early detection of cervical cancer in HPV infected women. The method shows promise in addressing the ongoing SELDI reproducibility issues.
Neural-Network-Based Smart Sensor Framework Operating in a Harsh Environment
Directory of Open Access Journals (Sweden)
Chaudhari Narendra S
2005-01-01
Full Text Available We present an artificial neural-network- (NN- based smart interface framework for sensors operating in harsh environments. The NN-based sensor can automatically compensate for the nonlinear response characteristics and its nonlinear dependency on the environmental parameters, with high accuracy. To show the potential of the proposed NN-based framework, we provide results of a smart capacitive pressure sensor (CPS operating in a wide temperature range of 0 to . Through simulated experiments, we have shown that the NN-based CPS model is capable of providing pressure readout with a maximum full-scale (FS error of only over this temperature range. A novel scheme for estimating the ambient temperature from the sensor characteristics itself is proposed. For this purpose, a second NN is utilized to estimate the ambient temperature accurately from the knowledge of the offset capacitance of the CPS. A microcontroller-unit- (MCU- based implementation scheme is also provided.
Kim, Junkyeong; Lee, Chaggil; Park, Seunghee
2017-06-07
Concrete is one of the most common materials used to construct a variety of civil infrastructures. However, since concrete might be susceptible to brittle fracture, it is essential to confirm the strength of concrete at the early-age stage of the curing process to prevent unexpected collapse. To address this issue, this study proposes a novel method to estimate the early-age strength of concrete, by integrating an artificial neural network algorithm with a dynamic response measurement of the concrete material. The dynamic response signals of the concrete, including both electromechanical impedances and guided ultrasonic waves, are obtained from an embedded piezoelectric sensor module. The cross-correlation coefficient of the electromechanical impedance signals and the amplitude of the guided ultrasonic wave signals are selected to quantify the variation in dynamic responses according to the strength of the concrete. Furthermore, an artificial neural network algorithm is used to verify a relationship between the variation in dynamic response signals and concrete strength. The results of an experimental study confirm that the proposed approach can be effectively applied to estimate the strength of concrete material from the early-age stage of the curing process.
Xiong, Wenjun; Patel, Ragini; Cao, Jinde; Zheng, Wei Xing
In this brief, our purpose is to apply asynchronous and intermittent sampled-data control methods to achieve the synchronization of hierarchical time-varying neural networks. The asynchronous and intermittent sampled-data controllers are proposed for two reasons: 1) the controllers may not transmit the control information simultaneously and 2) the controllers cannot always exist at any time . The synchronization is then discussed for a kind of hierarchical time-varying neural networks based on the asynchronous and intermittent sampled-data controllers. Finally, the simulation results are given to illustrate the usefulness of the developed criteria.In this brief, our purpose is to apply asynchronous and intermittent sampled-data control methods to achieve the synchronization of hierarchical time-varying neural networks. The asynchronous and intermittent sampled-data controllers are proposed for two reasons: 1) the controllers may not transmit the control information simultaneously and 2) the controllers cannot always exist at any time . The synchronization is then discussed for a kind of hierarchical time-varying neural networks based on the asynchronous and intermittent sampled-data controllers. Finally, the simulation results are given to illustrate the usefulness of the developed criteria.
Artificial neural networks-based approach to design ARIs using QSAR for diabetes mellitus.
Patra, Jagdish C; Singh, Onkar
2009-11-30
In this article, in the first part, we propose an artificial neural network-based intelligent technique to determine the quantitative structure-activity relationship (QSAR) among known aldose reductase inhibitors (ARIs) for diabetes mellitus using two molecular descriptors, i.e., the electronegativity and molar volume of functional groups present in the main ARI lead structure. We have shown that the multilayer perceptron-based model is capable of determining the QSAR quite satisfactorily, with high R-value. Usually, the design of potent ARIs requires the use of complex computer docking and quantum mechanical (QM) steps involving excessive time and human judgement. In the second part of this article, to reduce the design cycle of potent ARIs, we propose a novel ANN technique to eliminate the computer docking and QM steps, to predict the total score. The MLP-based QSAR models obtained in the first part are used to predict the potent ARIs, using the experimental data reported by Hu et al. (J Mol Graph Mod 2006, 24, 244). The proposed ANN-based model can predict the total score with an R-value of 0.88, which indicates that there exists a close match between the predicted and experimental total scores. Using the ANN model, we obtained 71 potent ARIs out of 6.25 million new ARI compounds created by substituting different functional groups at substituting sites of main lead structure of known ARI. Finally, using high bioactivity relationship and total score values, we determined four potential ARIs out of these 71 compounds. Interestingly, these four ARIs include the two potent ARIs reported by Hu et al. (J Mol Graph Mod 2006, 24, 244) who obtained these through the complex computer docking and QM steps. This fact indicates the effectiveness of our proposed ANN-based technique. We suggest these four compounds to be the most promising candidates for ARIs to prevent the diabetic complications and further recommend for wet bench experiments to find their potential against
Poultangari, Iman; Shahnazi, Reza; Sheikhan, Mansour
2012-09-01
In order to control the pitch angle of blades in wind turbines, commonly the proportional and integral (PI) controller due to its simplicity and industrial usability is employed. The neural networks and evolutionary algorithms are tools that provide a suitable ground to determine the optimal PI gains. In this paper, a radial basis function (RBF) neural network based PI controller is proposed for collective pitch control (CPC) of a 5-MW wind turbine. In order to provide an optimal dataset to train the RBF neural network, particle swarm optimization (PSO) evolutionary algorithm is used. The proposed method does not need the complexities, nonlinearities and uncertainties of the system under control. The simulation results show that the proposed controller has satisfactory performance. Copyright © 2012 ISA. Published by Elsevier Ltd. All rights reserved.
Quantum neural network-based EEG filtering for a brain-computer interface.
Gandhi, Vaibhav; Prasad, Girijesh; Coyle, Damien; Behera, Laxmidhar; McGinnity, Thomas Martin
2014-02-01
A novel neural information processing architecture inspired by quantum mechanics and incorporating the well-known Schrodinger wave equation is proposed in this paper. The proposed architecture referred to as recurrent quantum neural network (RQNN) can characterize a nonstationary stochastic signal as time-varying wave packets. A robust unsupervised learning algorithm enables the RQNN to effectively capture the statistical behavior of the input signal and facilitates the estimation of signal embedded in noise with unknown characteristics. The results from a number of benchmark tests show that simple signals such as dc, staircase dc, and sinusoidal signals embedded within high noise can be accurately filtered and particle swarm optimization can be employed to select model parameters. The RQNN filtering procedure is applied in a two-class motor imagery-based brain-computer interface where the objective was to filter electroencephalogram (EEG) signals before feature extraction and classification to increase signal separability. A two-step inner-outer fivefold cross-validation approach is utilized to select the algorithm parameters subject-specifically for nine subjects. It is shown that the subject-specific RQNN EEG filtering significantly improves brain-computer interface performance compared to using only the raw EEG or Savitzky-Golay filtered EEG across multiple sessions.
Joung, Semin; Kwak, Sehyun; Ghim, Y.-C.
2017-10-01
Obtaining plasma shapes during tokamak discharges requires real-time estimation of magnetic configuration using Grad-Shafranov solver such as EFIT. Since off-line EFIT is computationally intensive and the real-time reconstructions do not agree with the results of off-line EFIT within our desired accuracy, we use a neural network to generate an off-line-quality equilibrium in real time. To train the neural network (two hidden layers with 30 and 20 nodes for each layer), we create database consisting of the magnetic signals and off-line EFIT results from KSTAR as inputs and targets, respectively. To compensate drifts in the magnetic signals originated from electronic circuits, we develop a Bayesian-based two-step real-time correction method. Additionally, we infer missing inputs, i.e. when some of inputs to the network are not usable, using Gaussian process coupled with Bayesian model. The likelihood of this model is determined based on the Maxwell's equations. We find that our network can withstand at least up to 20% of input errors. Note that this real-time reconstruction scheme is not yet implemented for KSTAR operation.
Neural Network-Based Passive Filtering for Delayed Neutral-Type Semi-Markovian Jump Systems.
Shi, Peng; Li, Fanbiao; Wu, Ligang; Lim, Cheng-Chew
2017-09-01
This paper investigates the problem of exponential passive filtering for a class of stochastic neutral-type neural networks with both semi-Markovian jump parameters and mixed time delays. Our aim is to estimate the states by designing a Luenberger-type observer, such that the filter error dynamics are mean-square exponentially stable with an expected decay rate and an attenuation level. Sufficient conditions for the existence of passive filters are obtained, and a convex optimization algorithm for the filter design is given. In addition, a cone complementarity linearization procedure is employed to cast the nonconvex feasibility problem into a sequential minimization problem, which can be readily solved by the existing optimization techniques. Numerical examples are given to demonstrate the effectiveness of the proposed techniques.
Time-dependent prediction degredation assessment of neural-networks-based TEC forecasting models
Directory of Open Access Journals (Sweden)
Th. D. Xenos
2003-01-01
Full Text Available An estimation of the difference in TEC prediction accuracy achieved when the prediction varies from 1 h to 7 days in advance is described using classical neural networks. Hourly-daily Faraday-rotation derived TEC measurements from Florence are used. It is shown that the prediction accuracy for the examined dataset, though degrading when time span increases, is always high. In fact, when a relative prediction error margin of ± 10% is considered, the population percentage included therein is almost always well above the 55%. It is found that the results are highly dependent on season and the dataset wealth, whereas they highly depend on the foF2 - TEC variability difference and on hysteresis-like effect between these two ionospheric characteristics.
Data systems and computer science: Neural networks base R/T program overview
Gulati, Sandeep
1991-01-01
The research base, in the U.S. and abroad, for the development of neural network technology is discussed. The technical objectives are to develop and demonstrate adaptive, neural information processing concepts. The leveraging of external funding is also discussed.
Kucuk, Fuat; Goto, Hiroki; Guo, Hai-Jiao; Ichinokura, Osamu
2009-04-01
Feedback of motor torque is required in most of switched reluctance (SR) motor applications in order to control torque and its ripple. An SR motor shows highly nonlinear property which does not allow calculating torque analytically. Torque can be directly measured by torque sensor, but it inevitably increases the cost and has to be properly mounted on the motor shaft. Instead of torque sensor, finite element analysis (FEA) may be employed for torque calculation. However, motor modeling and calculation takes relatively long time. The results of FEA may also differ from the actual results. The most convenient way seems to calculate torque from the measured values of rotor position, current, and flux linkage while locking the rotor at definite positions. However, this method needs an extra assembly to lock the rotor. In this study, a novel torque calculation based on artificial neural networks (ANNs) is presented. Magnetizing data are collected while a 6/4 SR motor is running. They need to be interpolated for torque calculation. ANN is very strong tool for data interpolation. ANN based torque estimation is verified on the 6/4 SR motor and is compared by FEA based torque estimation to show its validity.
A Real Valued Neural Network Based Autoregressive Energy Detector for Cognitive Radio Application.
Onumanyi, A J; Onwuka, E N; Aibinu, A M; Ugweje, O C; Salami, M J E
2014-01-01
A real valued neural network (RVNN) based energy detector (ED) is proposed and analyzed for cognitive radio (CR) application. This was developed using a known two-layered RVNN model to estimate the model coefficients of an autoregressive (AR) system. By using appropriate modules and a well-designed detector, the power spectral density (PSD) of the AR system transfer function was estimated and subsequent receiver operating characteristic (ROC) curves of the detector generated and analyzed. A high detection performance with low false alarm rate was observed for varying signal to noise ratio (SNR), sample number, and model order conditions. The proposed RVNN based ED was then compared to the simple periodogram (SP), Welch periodogram (WP), multitaper (MT), Yule-Walker (YW), Burg (BG), and covariance (CV) based ED techniques. The proposed detector showed better performance than the SP, WP, and MT while providing better false alarm performance than the YW, BG, and CV. Data provided here support the effectiveness of the proposed RVNN based ED for CR application.
DEFF Research Database (Denmark)
Nielsen, Morten; Lund, Ole
2009-01-01
this binding event. RESULTS: Here, we present a novel artificial neural network-based method, NN-align that allows for simultaneous identification of the MHC class II binding core and binding affinity. NN-align is trained using a novel training algorithm that allows for correction of bias in the training data...... class II alleles, and is demonstrated to outperform other state-of-the-art MHC class II prediction methods. CONCLUSION: The NN-align method is competitive with the state-of-the-art MHC class II peptide binding prediction algorithms. The method is publicly available at http://www.cbs.dtu.dk/services/Net...
DEFF Research Database (Denmark)
Emanuelsson, O.; Nielsen, Henrik; von Heijne, Gunnar
1999-01-01
We present a neural network based method (ChloroP) for identifying chloroplast transit peptides and their cleavage sites. Using cross-validation, 88% of the sequences in our homology reduced training set were correctly classified as transit peptides or nontransit peptides. This performance level...... is well above that of the publicly available chloroplast localization predictor PSORT. Cleavage sites are predicted using a scoring matrix derived by an automatic motif-finding algorithm. Approximately 60% of the known cleavage sites in our sequence collection were predicted to within +/-2 residues from...
Directory of Open Access Journals (Sweden)
Lorena Álvarez
2009-01-01
Full Text Available Sound classifiers embedded in digital hearing aids are usually designed by using sound databases that do not include the distortions associated to the feedback that often occurs when these devices have to work at high gain and low gain margin to oscillation. The consequence is that the classifier learns inappropriate sound patterns. In this paper we explore the feasibility of using different sound databases (generated according to 18 configurations of real patients, and a variety of learning strategies for neural networks in the effort of reducing the probability of erroneous classification. The experimental work basically points out that the proposed methods assist the neural network-based classifier in reducing its error probability in more than 18%. This helps enhance the elderly user's comfort: the hearing aid automatically selects, with higher success probability, the program that is best adapted to the changing acoustic environment the user is facing.
Ferreira, Pedro M; Gomes, João M; Martins, Igor A C; Ruano, António E
2012-11-12
Accurate measurements of global solar radiation and atmospheric temperature,as well as the availability of the predictions of their evolution over time, are important for different areas of applications, such as agriculture, renewable energy and energy management, or thermal comfort in buildings. For this reason, an intelligent, light-weight and portable sensor was developed, using artificial neural network models as the time-series predictor mechanisms. These have been identified with the aid of a procedure based on the multi-objective genetic algorithm. As cloudiness is the most significant factor affecting the solar radiation reaching a particular location on the Earth surface, it has great impact on the performance of predictive solar radiation models for that location. This work also represents one step towards the improvement of such models by using ground-to-sky hemispherical colour digital images as a means to estimate cloudiness by the fraction of visible sky corresponding to clouds and to clear sky. The implementation of predictive models in the prototype has been validated and the system is able to function reliably, providing measurements and four-hour forecasts of cloudiness, solar radiation and air temperature.
Ferreira, Pedro M.; Gomes, João M.; Martins, Igor A. C.; Ruano, António E.
2012-01-01
Accurate measurements of global solar radiation and atmospheric temperature, as well as the availability of the predictions of their evolution over time, are important for different areas of applications, such as agriculture, renewable energy and energy management, or thermal comfort in buildings. For this reason, an intelligent, light-weight and portable sensor was developed, using artificial neural network models as the time-series predictor mechanisms. These have been identified with the aid of a procedure based on the multi-objective genetic algorithm. As cloudiness is the most significant factor affecting the solar radiation reaching a particular location on the Earth surface, it has great impact on the performance of predictive solar radiation models for that location. This work also represents one step towards the improvement of such models by using ground-to-sky hemispherical colour digital images as a means to estimate cloudiness by the fraction of visible sky corresponding to clouds and to clear sky. The implementation of predictive models in the prototype has been validated and the system is able to function reliably, providing measurements and four-hour forecasts of cloudiness, solar radiation and air temperature. PMID:23202230
Short-term load and wind power forecasting using neural network-based prediction intervals.
Quan, Hao; Srinivasan, Dipti; Khosravi, Abbas
2014-02-01
Electrical power systems are evolving from today's centralized bulk systems to more decentralized systems. Penetrations of renewable energies, such as wind and solar power, significantly increase the level of uncertainty in power systems. Accurate load forecasting becomes more complex, yet more important for management of power systems. Traditional methods for generating point forecasts of load demands cannot properly handle uncertainties in system operations. To quantify potential uncertainties associated with forecasts, this paper implements a neural network (NN)-based method for the construction of prediction intervals (PIs). A newly introduced method, called lower upper bound estimation (LUBE), is applied and extended to develop PIs using NN models. A new problem formulation is proposed, which translates the primary multiobjective problem into a constrained single-objective problem. Compared with the cost function, this new formulation is closer to the primary problem and has fewer parameters. Particle swarm optimization (PSO) integrated with the mutation operator is used to solve the problem. Electrical demands from Singapore and New South Wales (Australia), as well as wind power generation from Capital Wind Farm, are used to validate the PSO-based LUBE method. Comparative results show that the proposed method can construct higher quality PIs for load and wind power generation forecasts in a short time.
Directory of Open Access Journals (Sweden)
Bertin Dominique
2016-01-01
Full Text Available The Liane River is a small costal river, famous for its floods, which can affect the city of Boulogne-sur-Mer. Due to the complexity of land cover and hydrologic processes, a black-box non-linear modelling was chosen using neural networks. The multilayer perceptron model, known for its property of universal approximation is thus chosen. Four models were designed, each one for one forecasting horizon using rainfall forecasts: 24h, 12h, 6h, 3h. The desired output of the model is original: it represents the maximal value of the water level respectively 24h, 12h, 6h, 3h ahead. Working with best forecasts of rain (the observed ones during the event in the past, on the major flood of the database in test set, the model provides excellent forecasts. Nash criteria calculated for the four lead times are 0.98 (3h, 0.97 (6h, 0.91 (12h, 0.89 (24h. Designed models were thus estimated as efficient enough to be implemented in a specific tool devoted to real time operational use. The software tool is described hereafter: designed in Java, it presents a friendly interface allowing applying various scenarios of future rainfalls, and a graphical visualization of the predicted maximum water levels and their associated real time observed values.
Convolutional Neural Network-Based Robot Navigation Using Uncalibrated Spherical Images †
Ran, Lingyan; Zhang, Yanning; Zhang, Qilin; Yang, Tao
2017-01-01
Vision-based mobile robot navigation is a vibrant area of research with numerous algorithms having been developed, the vast majority of which either belong to the scene-oriented simultaneous localization and mapping (SLAM) or fall into the category of robot-oriented lane-detection/trajectory tracking. These methods suffer from high computational cost and require stringent labelling and calibration efforts. To address these challenges, this paper proposes a lightweight robot navigation framework based purely on uncalibrated spherical images. To simplify the orientation estimation, path prediction and improve computational efficiency, the navigation problem is decomposed into a series of classification tasks. To mitigate the adverse effects of insufficient negative samples in the “navigation via classification” task, we introduce the spherical camera for scene capturing, which enables 360° fisheye panorama as training samples and generation of sufficient positive and negative heading directions. The classification is implemented as an end-to-end Convolutional Neural Network (CNN), trained on our proposed Spherical-Navi image dataset, whose category labels can be efficiently collected. This CNN is capable of predicting potential path directions with high confidence levels based on a single, uncalibrated spherical image. Experimental results demonstrate that the proposed framework outperforms competing ones in realistic applications. PMID:28604624
Color matching of fabric blends: hybrid Kubelka-Munk + artificial neural network based method
Furferi, Rocco; Governi, Lapo; Volpe, Yary
2016-11-01
Color matching of fabric blends is a key issue for the textile industry, mainly due to the rising need to create high-quality products for the fashion market. The process of mixing together differently colored fibers to match a desired color is usually performed by using some historical recipes, skillfully managed by company colorists. More often than desired, the first attempt in creating a blend is not satisfactory, thus requiring the experts to spend efforts in changing the recipe with a trial-and-error process. To confront this issue, a number of computer-based methods have been proposed in the last decades, roughly classified into theoretical and artificial neural network (ANN)-based approaches. Inspired by the above literature, the present paper provides a method for accurate estimation of spectrophotometric response of a textile blend composed of differently colored fibers made of different materials. In particular, the performance of the Kubelka-Munk (K-M) theory is enhanced by introducing an artificial intelligence approach to determine a more consistent value of the nonlinear function relationship between the blend and its components. Therefore, a hybrid K-M+ANN-based method capable of modeling the color mixing mechanism is devised to predict the reflectance values of a blend.
Directory of Open Access Journals (Sweden)
Pedro M. Ferreira
2012-11-01
Full Text Available Accurate measurements of global solar radiation and atmospheric temperature,as well as the availability of the predictions of their evolution over time, are importantfor different areas of applications, such as agriculture, renewable energy and energymanagement, or thermal comfort in buildings. For this reason, an intelligent, light-weightand portable sensor was developed, using artificial neural network models as the time-seriespredictor mechanisms. These have been identified with the aid of a procedure based on themulti-objective genetic algorithm. As cloudiness is the most significant factor affecting thesolar radiation reaching a particular location on the Earth surface, it has great impact on theperformance of predictive solar radiation models for that location. This work also representsone step towards the improvement of such models by using ground-to-sky hemisphericalcolour digital images as a means to estimate cloudiness by the fraction of visible skycorresponding to clouds and to clear sky. The implementation of predictive models inthe prototype has been validated and the system is able to function reliably, providingmeasurements and four-hour forecasts of cloudiness, solar radiation and air temperature.
Neural network based tomographic approach to detect earthquake-related ionospheric anomalies
Directory of Open Access Journals (Sweden)
S. Hirooka
2011-08-01
Full Text Available A tomographic approach is used to investigate the fine structure of electron density in the ionosphere. In the present paper, the Residual Minimization Training Neural Network (RMTNN method is selected as the ionospheric tomography with which to investigate the detailed structure that may be associated with earthquakes. The 2007 Southern Sumatra earthquake (M = 8.5 was selected because significant decreases in the Total Electron Content (TEC have been confirmed by GPS and global ionosphere map (GIM analyses. The results of the RMTNN approach are consistent with those of TEC approaches. With respect to the analyzed earthquake, we observed significant decreases at heights of 250–400 km, especially at 330 km. However, the height that yields the maximum electron density does not change. In the obtained structures, the regions of decrease are located on the southwest and southeast sides of the Integrated Electron Content (IEC (altitudes in the range of 400–550 km and on the southern side of the IEC (altitudes in the range of 250–400 km. The global tendency is that the decreased region expands to the east with increasing altitude and concentrates in the Southern hemisphere over the epicenter. These results indicate that the RMTNN method is applicable to the estimation of ionospheric electron density.
Convolutional Neural Network-Based Robot Navigation Using Uncalibrated Spherical Images.
Ran, Lingyan; Zhang, Yanning; Zhang, Qilin; Yang, Tao
2017-06-12
Vision-based mobile robot navigation is a vibrant area of research with numerous algorithms having been developed, the vast majority of which either belong to the scene-oriented simultaneous localization and mapping (SLAM) or fall into the category of robot-oriented lane-detection/trajectory tracking. These methods suffer from high computational cost and require stringent labelling and calibration efforts. To address these challenges, this paper proposes a lightweight robot navigation framework based purely on uncalibrated spherical images. To simplify the orientation estimation, path prediction and improve computational efficiency, the navigation problem is decomposed into a series of classification tasks. To mitigate the adverse effects of insufficient negative samples in the "navigation via classification" task, we introduce the spherical camera for scene capturing, which enables 360° fisheye panorama as training samples and generation of sufficient positive and negative heading directions. The classification is implemented as an end-to-end Convolutional Neural Network (CNN), trained on our proposed Spherical-Navi image dataset, whose category labels can be efficiently collected. This CNN is capable of predicting potential path directions with high confidence levels based on a single, uncalibrated spherical image. Experimental results demonstrate that the proposed framework outperforms competing ones in realistic applications.
Convolutional Neural Network-Based Robot Navigation Using Uncalibrated Spherical Images
Directory of Open Access Journals (Sweden)
Lingyan Ran
2017-06-01
Full Text Available Vision-based mobile robot navigation is a vibrant area of research with numerous algorithms having been developed, the vast majority of which either belong to the scene-oriented simultaneous localization and mapping (SLAM or fall into the category of robot-oriented lane-detection/trajectory tracking. These methods suffer from high computational cost and require stringent labelling and calibration efforts. To address these challenges, this paper proposes a lightweight robot navigation framework based purely on uncalibrated spherical images. To simplify the orientation estimation, path prediction and improve computational efficiency, the navigation problem is decomposed into a series of classification tasks. To mitigate the adverse effects of insufficient negative samples in the “navigation via classification” task, we introduce the spherical camera for scene capturing, which enables 360° fisheye panorama as training samples and generation of sufficient positive and negative heading directions. The classification is implemented as an end-to-end Convolutional Neural Network (CNN, trained on our proposed Spherical-Navi image dataset, whose category labels can be efficiently collected. This CNN is capable of predicting potential path directions with high confidence levels based on a single, uncalibrated spherical image. Experimental results demonstrate that the proposed framework outperforms competing ones in realistic applications.
Boussalis, Dhemetrios; Wang, Shyh J.
1992-01-01
This paper presents a method for utilizing artificial neural networks for direct adaptive control of dynamic systems with poorly known dynamics. The neural network weights (controller gains) are adapted in real time using state measurements and a random search optimization algorithm. The results are demonstrated via simulation using two highly nonlinear systems.
PID Neural Network Based Speed Control of Asynchronous Motor Using Programmable Logic Controller
Directory of Open Access Journals (Sweden)
MARABA, V. A.
2011-11-01
Full Text Available This paper deals with the structure and characteristics of PID Neural Network controller for single input and single output systems. PID Neural Network is a new kind of controller that includes the advantages of artificial neural networks and classic PID controller. Functioning of this controller is based on the update of controller parameters according to the value extracted from system output pursuant to the rules of back propagation algorithm used in artificial neural networks. Parameters obtained from the application of PID Neural Network training algorithm on the speed model of the asynchronous motor exhibiting second order linear behavior were used in the real time speed control of the motor. Programmable logic controller (PLC was used as real time controller. The real time control results show that reference speed successfully maintained under various load conditions.
Megherbi, Dalila B.; Lodhi, S. M.; Boulenouar, A. J.
2001-03-01
This work is in the field of automated document processing. This work addresses the problem of representation and recognition of Urdu characters using Fourier representation and a Neural Network architecture. In particular, we show that a two-stage Neural Network scheme is used here to make classification of 36 Urdu characters into seven sub-classes namely subclasses characterized by seven proposed and defined fuzzy features specifically related to Urdu characters. We show that here Fourier Descriptors and Neural Network provide a remarkably simple way to draw definite conclusions from vague, ambiguous, noisy or imprecise information. In particular, we illustrate the concept of interest regions and describe a framing method that provides a way to make the proposed technique for Urdu characters recognition robust and invariant to scaling and translation. We also show that a given character rotation is dealt with by using the Hotelling transform. This transform is based upon the eigenvalue decomposition of the covariance matrix of an image, providing a method of determining the orientation of the major axis of an object within an image. Finally experimental results are presented to show the power and robustness of the proposed two-stage Neural Network based technique for Urdu character recognition, its fault tolerance, and high recognition accuracy.
Convolutional neural network-based data page classification for holographic memory.
Shimobaba, Tomoyoshi; Kuwata, Naoki; Homma, Mizuha; Takahashi, Takayuki; Nagahama, Yuki; Sano, Marie; Hasegawa, Satoki; Hirayama, Ryuji; Kakue, Takashi; Shiraki, Atsushi; Takada, Naoki; Ito, Tomoyoshi
2017-09-10
We propose a deep-learning-based classification of data pages used in holographic memory. We numerically investigated the classification performance of a conventional multilayer perceptron (MLP) and a deep neural network, under the condition that reconstructed page data are contaminated by some noise and are randomly laterally shifted. When data pages are randomly laterally shifted, the MLP was found to have a classification accuracy of 93.02%, whereas the deep neural network was able to classify data pages at an accuracy of 99.98%. The accuracy of the deep neural network is 2 orders of magnitude better than the MLP.
Neural Network-Based Self-Tuning PID Control for Underwater Vehicles.
Hernández-Alvarado, Rodrigo; García-Valdovinos, Luis Govinda; Salgado-Jiménez, Tomás; Gómez-Espinosa, Alfonso; Fonseca-Navarro, Fernando
2016-09-05
For decades, PID (Proportional + Integral + Derivative)-like controllers have been successfully used in academia and industry for many kinds of plants. This is thanks to its simplicity and suitable performance in linear or linearized plants, and under certain conditions, in nonlinear ones. A number of PID controller gains tuning approaches have been proposed in the literature in the last decades; most of them off-line techniques. However, in those cases wherein plants are subject to continuous parametric changes or external disturbances, online gains tuning is a desirable choice. This is the case of modular underwater ROVs (Remotely Operated Vehicles) where parameters (weight, buoyancy, added mass, among others) change according to the tool it is fitted with. In practice, some amount of time is dedicated to tune the PID gains of a ROV. Once the best set of gains has been achieved the ROV is ready to work. However, when the vehicle changes its tool or it is subject to ocean currents, its performance deteriorates since the fixed set of gains is no longer valid for the new conditions. Thus, an online PID gains tuning algorithm should be implemented to overcome this problem. In this paper, an auto-tune PID-like controller based on Neural Networks (NN) is proposed. The NN plays the role of automatically estimating the suitable set of PID gains that achieves stability of the system. The NN adjusts online the controller gains that attain the smaller position tracking error. Simulation results are given considering an underactuated 6 DOF (degrees of freedom) underwater ROV. Real time experiments on an underactuated mini ROV are conducted to show the effectiveness of the proposed scheme.
Neural Network-Based Self-Tuning PID Control for Underwater Vehicles
Hernández-Alvarado, Rodrigo; García-Valdovinos, Luis Govinda; Salgado-Jiménez, Tomás; Gómez-Espinosa, Alfonso; Fonseca-Navarro, Fernando
2016-01-01
For decades, PID (Proportional + Integral + Derivative)-like controllers have been successfully used in academia and industry for many kinds of plants. This is thanks to its simplicity and suitable performance in linear or linearized plants, and under certain conditions, in nonlinear ones. A number of PID controller gains tuning approaches have been proposed in the literature in the last decades; most of them off-line techniques. However, in those cases wherein plants are subject to continuous parametric changes or external disturbances, online gains tuning is a desirable choice. This is the case of modular underwater ROVs (Remotely Operated Vehicles) where parameters (weight, buoyancy, added mass, among others) change according to the tool it is fitted with. In practice, some amount of time is dedicated to tune the PID gains of a ROV. Once the best set of gains has been achieved the ROV is ready to work. However, when the vehicle changes its tool or it is subject to ocean currents, its performance deteriorates since the fixed set of gains is no longer valid for the new conditions. Thus, an online PID gains tuning algorithm should be implemented to overcome this problem. In this paper, an auto-tune PID-like controller based on Neural Networks (NN) is proposed. The NN plays the role of automatically estimating the suitable set of PID gains that achieves stability of the system. The NN adjusts online the controller gains that attain the smaller position tracking error. Simulation results are given considering an underactuated 6 DOF (degrees of freedom) underwater ROV. Real time experiments on an underactuated mini ROV are conducted to show the effectiveness of the proposed scheme. PMID:27608018
Neural Network-Based Self-Tuning PID Control for Underwater Vehicles
Directory of Open Access Journals (Sweden)
Rodrigo Hernández-Alvarado
2016-09-01
Full Text Available For decades, PID (Proportional + Integral + Derivative-like controllers have been successfully used in academia and industry for many kinds of plants. This is thanks to its simplicity and suitable performance in linear or linearized plants, and under certain conditions, in nonlinear ones. A number of PID controller gains tuning approaches have been proposed in the literature in the last decades; most of them off-line techniques. However, in those cases wherein plants are subject to continuous parametric changes or external disturbances, online gains tuning is a desirable choice. This is the case of modular underwater ROVs (Remotely Operated Vehicles where parameters (weight, buoyancy, added mass, among others change according to the tool it is fitted with. In practice, some amount of time is dedicated to tune the PID gains of a ROV. Once the best set of gains has been achieved the ROV is ready to work. However, when the vehicle changes its tool or it is subject to ocean currents, its performance deteriorates since the fixed set of gains is no longer valid for the new conditions. Thus, an online PID gains tuning algorithm should be implemented to overcome this problem. In this paper, an auto-tune PID-like controller based on Neural Networks (NN is proposed. The NN plays the role of automatically estimating the suitable set of PID gains that achieves stability of the system. The NN adjusts online the controller gains that attain the smaller position tracking error. Simulation results are given considering an underactuated 6 DOF (degrees of freedom underwater ROV. Real time experiments on an underactuated mini ROV are conducted to show the effectiveness of the proposed scheme.
Real-time neural network-based self-tuning control of a nonlinear electro-hydraulic servomotor
Energy Technology Data Exchange (ETDEWEB)
Canelon, J.I.; Ortega, A.G. [Univ. del Zulia, Maracaibo, Zulia (Venezuela, Bolivarian Republic of). School of Electrical Engineering; Shieh, L.S. [Houston Univ., Houston, TX (United States). Dept. of Electrical and Computer Engineering; Bastidas, J.I. [Univ. del Zulia, Maracaibo, Zulia (Venezuela, Bolivarian Republic of). School of Mechanical Engineering; Zhang, Y.; Akujuobi, C.M. [Prairie View A and M Univ., Prairie View, TX (United States). Center of Excellence for Communication Systems Technology Research and Dept. of Engineering Technology
2010-08-13
For high power applications, hydraulic actuators offer many advantages over electromagnetic actuators, including higher torque/mass ratios; smaller control gains; excellent torque capability; filtered high frequency noise; better heat transfer characteristics; smaller size; higher speed of response of the servomechanism; cheaper hardware; and higher reliability. Therefore, any application that requires a large force applied smoothly by an actuator is a candidate for hydraulic power. Examples of such applications include vehicle steering and braking systems; roll mills; drilling rigs; heavy duty crane and presses; and industrial robots and actuators for aircraft control surfaces such as ailerons and flaps. It is extremely important to create effective control strategies for hydraulic systems. This paper outlined the real-time implementation of a neural network-based approach, for self-tuning control of the angular position of a nonlinear electro-hydraulic servomotor. Using an online training algorithm, a neural network autoregressive moving-average model with exogenous input (ARMAX) model of the system was identified and continuously updated and an optimal linear ARMAX model was determined. The paper briefly depicted the neural network-based self-tuning control approach and a description of the experimental equipment (hardware and software) was presented including the implementation details. The experimental results were discussed and conclusions were summarized. It was found that the approach proved to be very effective in the control of this fast dynamics system, outperforming a fine tuned PI controller. Therefore, although the self-tuning approach was computationally demanding, it was feasible for real-time implementation. 22 refs., 6 figs.
Estimation of Conditional Quantile using Neural Networks
DEFF Research Database (Denmark)
Kulczycki, P.; Schiøler, Henrik
1999-01-01
The problem of estimating conditional quantiles using neural networks is investigated here. A basic structure is developed using the methodology of kernel estimation, and a theory guaranteeing con-sistency on a mild set of assumptions is provided. The constructed structure constitutes a basis...... for the design of a variety of different neural networks, some of which are considered in detail. The task of estimating conditional quantiles is related to Bayes point estimation whereby a broad range of applications within engineering, economics and management can be suggested. Numerical results illustrating...... the capabilities of the elaborated neural network are also given....
Directory of Open Access Journals (Sweden)
G. Olivito
2017-01-01
In the present study, the network-based statistics (NBS approach was used to assess differences in functional connectivity between specific cerebellar and cerebral “nodes” in SCA2 patients. Altered inter-nodal connectivity was found between more posterior regions in the cerebellum and regions in the cerebral cortex clearly related to cognition and emotion. Furthermore, more anterior cerebellar lobules showed altered inter-nodal connectivity with motor and somatosensory cerebral regions. The present data suggest that in SCA2 a cerebellar dysfunction affects long-distance cerebral regions and that the clinical symptoms may be specifically related with connectivity changes between motor and non-motor cerebello-cortical nodes.
Single-hidden-layer feed-forward quantum neural network based on Grover learning.
Liu, Cheng-Yi; Chen, Chein; Chang, Ching-Ter; Shih, Lun-Min
2013-09-01
In this paper, a novel single-hidden-layer feed-forward quantum neural network model is proposed based on some concepts and principles in the quantum theory. By combining the quantum mechanism with the feed-forward neural network, we defined quantum hidden neurons and connected quantum weights, and used them as the fundamental information processing unit in a single-hidden-layer feed-forward neural network. The quantum neurons make a wide range of nonlinear functions serve as the activation functions in the hidden layer of the network, and the Grover searching algorithm outstands the optimal parameter setting iteratively and thus makes very efficient neural network learning possible. The quantum neuron and weights, along with a Grover searching algorithm based learning, result in a novel and efficient neural network characteristic of reduced network, high efficient training and prospect application in future. Some simulations are taken to investigate the performance of the proposed quantum network and the result show that it can achieve accurate learning. Copyright © 2013 Elsevier Ltd. All rights reserved.
Neural network-based adaptive dynamic surface control for permanent magnet synchronous motors.
Yu, Jinpeng; Shi, Peng; Dong, Wenjie; Chen, Bing; Lin, Chong
2015-03-01
This brief considers the problem of neural networks (NNs)-based adaptive dynamic surface control (DSC) for permanent magnet synchronous motors (PMSMs) with parameter uncertainties and load torque disturbance. First, NNs are used to approximate the unknown and nonlinear functions of PMSM drive system and a novel adaptive DSC is constructed to avoid the explosion of complexity in the backstepping design. Next, under the proposed adaptive neural DSC, the number of adaptive parameters required is reduced to only one, and the designed neural controllers structure is much simpler than some existing results in literature, which can guarantee that the tracking error converges to a small neighborhood of the origin. Then, simulations are given to illustrate the effectiveness and potential of the new design technique.
Pennington, Robert S.; Coll, Catalina; Estradé, Sònia; Peiró, Francesca; Koch, Christoph T.
2018-01-01
Iterative neural-network-based three-dimensional structural optimization of atomic positions over tens of nanometers is performed using transmission electron microscope (TEM) diffraction data simulated from density functional theory (DFT) all-electron densities, thus retrieving parameter variations along the beam direction. We first use experimental data to show that the GPAW DFT code's all-electron densities are considerably more accurate for electron diffraction calculations compared to conventional isolated-atom scattering factors, and they also compare well to Wien2K DFT simulations. This DFT-TEM combination is then integrated into an iterative neural-network-optimization-based algorithm (PRIMES, parameter retrieval and inversion from multiple electron scattering) to retrieve nanometer-scale ferroelectric polarization domains and strain in theoretical bulklike specimens from TEM data. DFT and isolated-atom methods produce substantially different diffraction patterns and retrieved polarization domain parameters, and DFT is sufficient to retrieve strain properties from a silicon specimen simulated using experimentally derived structure factors. Thus, we show that the improved accuracy, fast computation, and intuitive integration make the GPAW DFT code well suited for three-dimensional materials characterization and demonstrate this using an iterative neural-network algorithm that is verifiable on the mesoscale and, with DFT integration, self-consistent on the nanoscale.
An Artificial Neural Network Based Short-term Dynamic Prediction of Algae Bloom
Directory of Open Access Journals (Sweden)
Yao Junyang
2014-06-01
Full Text Available This paper proposes a method of short-term prediction of algae bloom based on artificial neural network. Firstly, principal component analysis is applied to water environmental factors in algae bloom raceway ponds to get main factors that influence the formation of algae blooms. Then, a model of short-term dynamic prediction based on neural network is built with the current chlorophyll_a values as input and the chlorophyll_a values in the next moment as output to realize short-term dynamic prediction of algae bloom. Simulation results show that the model can realize short-term prediction of algae bloom effectively.
Neural Network Based Reactive Navigation for Mobile Robot in Dynamic Environment
Czech Academy of Sciences Publication Activity Database
Krejsa, Jiří; Věchet, S.; Ripel, T.
2013-01-01
Roč. 198, č. 2013 (2013), s. 108-113 ISSN 1012-0394 Institutional research plan: CEZ:AV0Z20760514 Institutional support: RVO:61388998 Keywords : mobile robot * reactive navigation * artificial neural networks Subject RIV: JD - Computer Applications, Robot ics
Neural networks-based modeling applied to a process of heavy metals removal from wastewaters.
Suditu, Gabriel D; Curteanu, Silvia; Bulgariu, Laura
2013-01-01
This article approaches the problem of environment pollution with heavy metals from disposal of industrial wastewaters, namely removal of these metals by means of biosorbents, particularly with Romanian peat (from Poiana Stampei). The study is carried out by simulation using feed-forward and modular neural networks with one or two hidden layers, pursuing the influence of certain operating parameters (metal nature, sorbent dose, pH, temperature, initial concentration of metal ion, contact time) on the amount of metal ions retained on the unit mass of sorbent. In neural network modeling, a consistent data set was used, including five metals: lead, mercury, cadmium, nickel and cobalt, the quantification of the metal nature being done by its electronegativity. Even if based on successive trials, the method of designing neural models was systematically conducted, recording and comparing the errors obtained with different types of neural networks, having various numbers of hidden layers and neurons, number of training epochs, or using various learning methods. The errors with values under 5% make clear the efficiency of the applied method.
CSIR Research Space (South Africa)
Ngwangwa, HM
2008-07-01
Full Text Available on the road and driver to assess the integrity of road and vehicle infrastructure. In this paper, vehicle vibration data are applied to an artificial neural network to reconstruct the corresponding road surface profiles. The results show that the technique...
neural network based model o work based model of an industrial oil
African Journals Online (AJOL)
eobe
ropagation Algorithm (trainlm). The data; shows that data; shows that the simulated model the simulated model t outputs t outputs, with regression val with regression val with regression value of d to control the neural network model. , PID controller. ing processes. An additional node with ut (usually 1) is often added to the.
Valdes, A.; Khorasani, K.
The main objective of this paper is to develop a dynamic neural network-based fault detection and isolation (FDI) scheme for the Pulsed Plasma Thrusters (PPTs) that are used in the Attitude Control Subsystem (ACS) of satellites that are tasked to perform a formation flying mission. By using data collected from the relative attitudes of the formation flying satellites our proposed "High Level" FDI scheme can detect the pair of thrusters which is faulty, however fault isolation cannot be accomplished. Based on the "High Level" FDI scheme and the DNN-based "Low Level" FDI scheme developed earlier by the authors, an "Integrated" DNN-based FDI scheme is then proposed. To demonstrate the FDI capabilities of the proposed schemes various fault scenarios are simulated.
Directory of Open Access Journals (Sweden)
Anh Vu Le
2017-01-01
Full Text Available In this paper, the depth images and the colour images provided by Kinect sensors are used to enhance the accuracy of human detection. The depth-based human detection method is fast but less accurate. On the other hand, the faster region convolutional neural network-based human detection method is accurate but requires a rather complex hardware configuration. To simultaneously leverage the advantages and relieve the drawbacks of each method, one master and one client system is proposed. The final goal is to make a novel Robot Operation System (ROS-based Perception Sensor Network (PSN system, which is more accurate and ready for the real time application. The experimental results demonstrate the outperforming of the proposed method compared with other conventional methods in the challenging scenarios.
Hu, Bin; Yue, Shigang; Zhang, Zhuhong
All complex motion patterns can be decomposed into several elements, including translation, expansion/contraction, and rotational motion. In biological vision systems, scientists have found that specific types of visual neurons have specific preferences to each of the three motion elements. There are computational models on translation and expansion/contraction perceptions; however, little has been done in the past to create computational models for rotational motion perception. To fill this gap, we proposed a neural network that utilizes a specific spatiotemporal arrangement of asymmetric lateral inhibited direction selective neural networks (DSNNs) for rotational motion perception. The proposed neural network consists of two parts-presynaptic and postsynaptic parts. In the presynaptic part, there are a number of lateral inhibited DSNNs to extract directional visual cues. In the postsynaptic part, similar to the arrangement of the directional columns in the cerebral cortex, these direction selective neurons are arranged in a cyclic order to perceive rotational motion cues. In the postsynaptic network, the delayed excitation from each direction selective neuron is multiplied by the gathered excitation from this neuron and its unilateral counterparts depending on which rotation, clockwise (cw) or counter-cw (ccw), to perceive. Systematic experiments under various conditions and settings have been carried out and validated the robustness and reliability of the proposed neural network in detecting cw or ccw rotational motion. This research is a critical step further toward dynamic visual information processing.All complex motion patterns can be decomposed into several elements, including translation, expansion/contraction, and rotational motion. In biological vision systems, scientists have found that specific types of visual neurons have specific preferences to each of the three motion elements. There are computational models on translation and expansion
An artificial neural network based $b$ jet identification algorithm at the CDF Experiment
Freeman, J; Ketchum, W; Poprocki, S; Pronko, A; Rusu, V; Wittich, P
2011-01-01
We present the development and validation of a new multivariate $b$ jet identification algorithm ("$b$ tagger") used at the CDF experiment at the Fermilab Tevatron. At collider experiments, $b$ taggers allow one to distinguish particle jets containing $B$ hadrons from other jets. Employing feed-forward neural network architectures, this tagger is unique in its emphasis on using information from individual tracks. This tagger not only contains the usual advantages of a multivariate technique such as maximal use of information in a jet and tunable purity/efficiency operating points, but is also capable of evaluating jets with only a single track. To demonstrate the effectiveness of the tagger, we employ a novel method wherein we calculate the false tag rate and tag efficiency as a function of the placement of a lower threshold on a jet's neural network output value in $Z+1$ jet and $t\\bar{t}$ candidate samples, rich in light flavor and $b$ jets, respectively.
Neural network-based survey analysis of risk management practices in new product development
DEFF Research Database (Denmark)
Kampianakis, Andreas N.; Oehmen, Josef
2017-01-01
The current study investigates the applicability of Artificial Neural Networks (ANNs) to analyse survey data on the effectiveness of risk management practices in product development (PD) projects, and its ability to forecast project outcomes. Moreover, this study presents the relations between risk...... Neural Networks. Dataset used is a filtered survey of 291 product development programs. Answers of this survey are used as training input and target output, in pattern recognition two-layer feed forward networks, using various transfer functions. Using this method, relations among 6 project practices...... and 13 outcome metrics were revealed. Results of this analysis are compared with existent results made through statistical analysis in prior work of one of the authors. Future investigation is needed in order to tackle the lack of data and create an easy to use platform for industrial use....
Classification of Weed Species Using Artificial Neural Networks Based on Color Leaf Texture Feature
Li, Zhichen; An, Qiu; Ji, Changying
The potential impact of herbicide utilization compel people to use new method of weed control. Selective herbicide application is optimal method to reduce herbicide usage while maintain weed control. The key of selective herbicide is how to discriminate weed exactly. The HIS color co-occurrence method (CCM) texture analysis techniques was used to extract four texture parameters: Angular second moment (ASM), Entropy(E), Inertia quadrature (IQ), and Inverse difference moment or local homogeneity (IDM).The weed species selected for studying were Arthraxon hispidus, Digitaria sanguinalis, Petunia, Cyperus, Alternanthera Philoxeroides and Corchoropsis psilocarpa. The software of neuroshell2 was used for designing the structure of the neural network, training and test the data. It was found that the 8-40-1 artificial neural network provided the best classification performance and was capable of classification accuracies of 78%.
Directory of Open Access Journals (Sweden)
Heryanto M Ary
2015-01-01
Full Text Available UAVs are mostly used for surveillance, inspection and data acquisition. We have developed a Quadrotor UAV that is constructed based on a four motors with a lift-generating propeller at each motors. In this paper, we discuss the development of a quadrotor and its neural networks direct inverse control model using the actual flight data. To obtain a better performance of the control system of the UAV, we proposed an Optimized Direct Inverse controller based on re-training the neural networks with the new data generated from optimal maneuvers of the quadrotor. Through simulation of the quadrotor using the developed DIC and Optimized DIC model, results show that both models have the ability to stabilize the quadrotor with a good tracking performance. The optimized DIC model, however, has shown a better performance, especially in the settling time parameter.
A neural network-based electromyography motion classifier for upper limb activities
Directory of Open Access Journals (Sweden)
Karan Veer
2016-11-01
Full Text Available The objective of the work is to investigate the classification of different movements based on the surface electromyogram (SEMG pattern recognition method. The testing was conducted for four arm movements using several experiments with artificial neural network classification scheme. Six time domain features were extracted and consequently classification was implemented using back propagation neural classifier (BPNC. Further, the realization of projected network was verified using cross validation (CV process; hence ANOVA algorithm was carried out. Performance of the network is analyzed by considering mean square error (MSE value. A comparison was performed between the extracted features and back propagation network results reported in the literature. The concurrent result indicates the significance of proposed network with classification accuracy (CA of 100% recorded from two channels, while analysis of variance technique helps in investigating the effectiveness of classified signal for recognition tasks.
Kolla, Sri R; Altman, Shawn D
2007-04-01
This paper presents results from the implementation and testing of a PC based monitoring and fault identification scheme for a three-phase induction motor using artificial neural networks (ANNs). To accomplish the task, a hardware system is designed and built to acquire three-phase voltages and currents from a 1/3 HP squirrel-cage, three-phase induction motor. A software program is written to read the voltages and currents, which are first used to train a feed-forward neural network structure using the JavaNNS program. The trained network is placed in a LabVIEW based program formula node that monitors the voltages and currents online and displays the fault conditions and turns the motor off. The complete system is successfully tested in real time by creating different faults on the motor.
Chaotic Extension Neural Network-Based Fault Diagnosis Method for Solar Photovoltaic Systems
Directory of Open Access Journals (Sweden)
Kuo-Nan Yu
2014-01-01
Full Text Available At present, the solar photovoltaic system is extensively used. However, once a fault occurs, it is inspected manually, which is not economical. In order to remedy the defect of unavailable fault diagnosis at any irradiance and temperature in the literature with chaos synchronization based intelligent fault diagnosis for photovoltaic systems proposed by Hsieh et al., this study proposed a chaotic extension fault diagnosis method combined with error back propagation neural network to overcome this problem. It used the nn toolbox of matlab 2010 for simulation and comparison, measured current irradiance and temperature, and used the maximum power point tracking (MPPT for chaotic extraction of eigenvalue. The range of extension field was determined by neural network. Finally, the voltage eigenvalue obtained from current temperature and irradiance was used for the fault diagnosis. Comparing the diagnostic rates with the results by Hsieh et al., this scheme can obtain better diagnostic rates when the irradiances or the temperatures are changed.
A Review of Neural Network Based Machine Learning Approaches for Rotor Angle Stability Control
Yousefian, Reza; Kamalasadan, Sukumar
2017-01-01
This paper reviews the current status and challenges of Neural Networks (NNs) based machine learning approaches for modern power grid stability control including their design and implementation methodologies. NNs are widely accepted as Artificial Intelligence (AI) approaches offering an alternative way to control complex and ill-defined problems. In this paper various application of NNs for power system rotor angle stabilization and control problem is discussed. The main focus of this paper i...
Vinodhini, G.; Chandrasekaran, R.M.
2016-01-01
The aim of sentiment classification is to efficiently identify the emotions expressed in the form of text messages. Machine learning methods for sentiment classification have been extensively studied, due to their predominant classification performance. Recent studies suggest that ensemble based machine learning methods provide better performance in classification. Artificial neural networks (ANNs) are rarely being investigated in the literature of sentiment classification. This paper compare...
Directory of Open Access Journals (Sweden)
Jilin Zhang
2017-01-01
Full Text Available With the development of the mobile systems, we gain a lot of benefits and convenience by leveraging mobile devices; at the same time, the information gathered by smartphones, such as location and environment, is also valuable for business to provide more intelligent services for customers. More and more machine learning methods have been used in the field of mobile information systems to study user behavior and classify usage patterns, especially convolutional neural network. With the increasing of model training parameters and data scale, the traditional single machine training method cannot meet the requirements of time complexity in practical application scenarios. The current training framework often uses simple data parallel or model parallel method to speed up the training process, which is why heterogeneous computing resources have not been fully utilized. To solve these problems, our paper proposes a delay synchronization convolutional neural network parallel strategy, which leverages the heterogeneous system. The strategy is based on both synchronous parallel and asynchronous parallel approaches; the model training process can reduce the dependence on the heterogeneous architecture in the premise of ensuring the model convergence, so the convolution neural network framework is more adaptive to different heterogeneous system environments. The experimental results show that the proposed delay synchronization strategy can achieve at least three times the speedup compared to the traditional data parallelism.
Kumar, Rajesh; Srivastava, Smriti; Gupta, J R P
2017-03-01
In this paper adaptive control of nonlinear dynamical systems using diagonal recurrent neural network (DRNN) is proposed. The structure of DRNN is a modification of fully connected recurrent neural network (FCRNN). Presence of self-recurrent neurons in the hidden layer of DRNN gives it an ability to capture the dynamic behaviour of the nonlinear plant under consideration (to be controlled). To ensure stability, update rules are developed using lyapunov stability criterion. These rules are then used for adjusting the various parameters of DRNN. The responses of plants obtained with DRNN are compared with those obtained when multi-layer feed forward neural network (MLFFNN) is used as a controller. Also, in example 4, FCRNN is also investigated and compared with DRNN and MLFFNN. Robustness of the proposed control scheme is also tested against parameter variations and disturbance signals. Four simulation examples including one-link robotic manipulator and inverted pendulum are considered on which the proposed controller is applied. The results so obtained show the superiority of DRNN over MLFFNN as a controller. Copyright © 2017 ISA. Published by Elsevier Ltd. All rights reserved.
Laidi, Maamar; Hanini, Salah; Rezrazi, Ahmed; Yaiche, Mohamed Redha; El Hadj, Abdallah Abdallah; Chellali, Farouk
2017-04-01
In this study, a backpropagation artificial neural network (BP-ANN) model is used as an alternative approach to predict solar radiation on tilted surfaces (SRT) using a number of variables involved in physical process. These variables are namely the latitude of the site, mean temperature and relative humidity, Linke turbidity factor and Angstrom coefficient, extraterrestrial solar radiation, solar radiation data measured on horizontal surfaces (SRH), and solar zenith angle. Experimental solar radiation data from 13 stations spread all over Algeria around the year (2004) were used for training/validation and testing the artificial neural networks (ANNs), and one station was used to make the interpolation of the designed ANN. The ANN model was trained, validated, and tested using 60, 20, and 20 % of all data, respectively. The configuration 8-35-1 (8 inputs, 35 hidden, and 1 output neurons) presented an excellent agreement between the prediction and the experimental data during the test stage with determination coefficient of 0.99 and root meat squared error of 5.75 Wh/m2, considering a three-layer feedforward backpropagation neural network with Levenberg-Marquardt training algorithm, a hyperbolic tangent sigmoid and linear transfer function at the hidden and the output layer, respectively. This novel model could be used by researchers or scientists to design high-efficiency solar devices that are usually tilted at an optimum angle to increase the solar incident on the surface.
Pandey, Daya Shankar; Das, Saptarshi; Pan, Indranil; Leahy, James J; Kwapinski, Witold
2016-12-01
In this paper, multi-layer feed forward neural networks are used to predict the lower heating value of gas (LHV), lower heating value of gasification products including tars and entrained char (LHVp) and syngas yield during gasification of municipal solid waste (MSW) during gasification in a fluidized bed reactor. These artificial neural networks (ANNs) with different architectures are trained using the Levenberg-Marquardt (LM) back-propagation algorithm and a cross validation is also performed to ensure that the results generalise to other unseen datasets. A rigorous study is carried out on optimally choosing the number of hidden layers, number of neurons in the hidden layer and activation function in a network using multiple Monte Carlo runs. Nine input and three output parameters are used to train and test various neural network architectures in both multiple output and single output prediction paradigms using the available experimental datasets. The model selection procedure is carried out to ascertain the best network architecture in terms of predictive accuracy. The simulation results show that the ANN based methodology is a viable alternative which can be used to predict the performance of a fluidized bed gasifier. Copyright © 2016 Elsevier Ltd. All rights reserved.
FPGA implementation of a biological neural network based on the Hodgkin-Huxley neuron model.
Yaghini Bonabi, Safa; Asgharian, Hassan; Safari, Saeed; Nili Ahmadabadi, Majid
2014-01-01
A set of techniques for efficient implementation of Hodgkin-Huxley-based (H-H) model of a neural network on FPGA (Field Programmable Gate Array) is presented. The central implementation challenge is H-H model complexity that puts limits on the network size and on the execution speed. However, basics of the original model cannot be compromised when effect of synaptic specifications on the network behavior is the subject of study. To solve the problem, we used computational techniques such as CORDIC (Coordinate Rotation Digital Computer) algorithm and step-by-step integration in the implementation of arithmetic circuits. In addition, we employed different techniques such as sharing resources to preserve the details of model as well as increasing the network size in addition to keeping the network execution speed close to real time while having high precision. Implementation of a two mini-columns network with 120/30 excitatory/inhibitory neurons is provided to investigate the characteristic of our method in practice. The implementation techniques provide an opportunity to construct large FPGA-based network models to investigate the effect of different neurophysiological mechanisms, like voltage-gated channels and synaptic activities, on the behavior of a neural network in an appropriate execution time. Additional to inherent properties of FPGA, like parallelism and re-configurability, our approach makes the FPGA-based system a proper candidate for study on neural control of cognitive robots and systems as well.
A neural network based model to analyze rice parboiling process with small dataset.
Behroozi-Khazaei, Nasser; Nasirahmadi, Abozar
2017-07-01
In this study, milling recovery, head rice yield, degree of milling and whiteness were utilized to characterize the milling quality of Tarom parboiled rice variety. The parboiled rice was prepared with three soaking temperatures and steaming times. Then the samples were dried to three levels of final moisture contents [8, 10 and 12% (w.b)]. Modeling of process and validating of the results with small dataset are always challenging. So, the aim of this study was to develop models based on the milling quality data in parboiling process by means of multivariate regression and artificial neural network. In order to validate the neural network model with a little dataset, K-fold cross validation method was applied. The ANN structure with one hidden layer and Tansig transfer function by 18 neurons in the hidden layer was selected as the best model in this study. The results indicated that the neural network could model the parboiling process with higher degree of accuracy. This method was a promising procedure to create accuracy and can be used as a reliable model to select the best parameters for the parboiling process with little experiment dataset.
Spiking Neural Networks based on OxRAM Synapses for Real-time Unsupervised Spike Sorting
Directory of Open Access Journals (Sweden)
Thilo Werner
2016-11-01
Full Text Available In this paper, we present an alternative approach to perform spike sorting of complex brain signals based on spiking neural networks (SNN. The proposed architecture is suitable for hardware implementation by using RRAM technology for the implementation of synapses whose low latency (< 1μs enable real-time spike sorting. This offers promising advantagesto conventional spike sorting techniques for brain-computer interface and neural prosthesis applications. Moreover, the ultralow power consumption of the RRAM synapses of the spiking neural network (nW range may enable the design of autonomous implantable devices for rehabilitation purposes. We demonstrate an original methodology to use Oxide based RRAM (OxRAM as easy to program and low power (< 75 pJ synapses. Synaptic weights are modulated through the application of an online learning strategy inspired by biological Spike Timing Dependent Plasticity. Real spiking data have been recorded both intraand extracellularly from an in-vitro preparation of the Crayfish sensory-motor system and used for validation of the proposed OxRAM based SNN. This artificial SNN is able to identify, learn, recognize and distinguish between different spike shapes in the input signal with a recognition rate about 90% without any supervision.
Detection of lung injury with conventional and neural network-based analysis of continuous data.
Räsänen, J; León, M A
1998-08-01
To test if analysis of pressure and flow waveform patterns with an artificial intelligence neural network could distinguish between normal and injured lungs. Acute lung injury was induced in ten healthy anesthetized, mechanically ventilated dogs with repeated injections of oleic acid, until arterial blood oxyhemoglobin saturation reached 85% breathing room air. Airway pressure, esophageal pressure, airway flow, and arterial and mixed venous saturation signals were stored at 2 min intervals. Hemodynamic and blood gas data were collected every 10 min. Back-propagation neural networks were trained with normalized airway pressure and flow waveforms from normal and fully injured lungs. The networks scored lung injury on a continuous scale from +1 (normal) to -1 (injured). Network scores unequivocally distinguished between normal and fully injured lungs and suggested a gradual transition from normal to injury pattern. However, the response of the network was slow compared to compliance, resistance and venous admixture. Normal and fully injured lungs display distinct flow and pressure waveform patterns which are independent of changes in calculated pulmonary mechanics variables. These patterns can be recognized by a neural network. Further research is needed to determine the full potential of automated pattern recognition for lung monitoring.
Directory of Open Access Journals (Sweden)
Maryam Rahmati Garousi
2016-03-01
Full Text Available In this paper, a new method for the detection and classification of all types of power quality disturbances is presented. In addition to separating the disturbance signals, the proposed method is able to determine the type of disturbances. Disturbance waveforms are generated based on IEEE 1159 standard and they are de-noised using discrete wavelet transform. To detect the sinusoidal signals from disturbance signals, new criteria have been proposed. By introducing these new criteria, the classification algorithm is not active for non-disturbance signals. Therefore, the computation time is reduced. If a signal has disturbance, to extract the required information, it is analyzed using discrete wavelet transform. Using this information, the appropriate feature vectors are introduced. Parallel neural networks structures are proposed for the classification of disturbances. The inputs of these networks are the introduced feature vectors. The proposed method is done for all power quality disturbances including DC offset, flicker, interrupt, sag, swell, harmonic, notching, impulsive transient, oscillatory transient and eight combinations of these including the harmonics with transient, harmonic with flicker, harmonic with sag, harmonic with swell, sag with flicker, swell with flicker, transient with sag and transient with swell. The performance of this algorithm is compared with a single neural network structure. The results indicate using the parallel neural networks structure, computational time is much reduced and the accuracy of classification of power quality disturbances is significantly increased. Comparison the obtained results by the method with other methods, represents very high performance of the proposed method with precision %99.53.
Neural-network-based single-sided non-enwrapping power loss tester
Passadis, K; Beckley, P
2003-01-01
It is preferable to be able to assess the power loss of electrical steels during production. When a single-sided tester is used, flux sensing is undertaken from one side only and hence some leakage flux above the strip may not captured by the sensing coils. Therefore, the disadvantage of a single-sided non-enwrapping tester lies in the measurement of the flux density in the material. A neural network was successfully used to 'predict' the correct level of flux density for accurate assessment of power loss.
Neural Network Based Recognition of Signal Patterns in Application to Automatic Testing of Rails
Directory of Open Access Journals (Sweden)
Tomasz Ciszewski
2006-01-01
Full Text Available The paper describes the application of neural network for recognition of signal patterns in measuring data gathered by the railroad ultrasound testing car. Digital conversion of the measuring signal allows to store and process large quantities of data. The elaboration of smart, effective and automatic procedures recognizing the obtained patterns on the basisof measured signal amplitude has been presented. The test shows only two classes of pattern recognition. In authors’ opinion if we deliver big enough quantity of training data, presented method is applicable to a system that recognizes many classes.
Directory of Open Access Journals (Sweden)
Yuxian Zhang
2015-01-01
Full Text Available The quality index model in slashing process is difficult to build by reason of the outliers and noise data from original data. To the above problem, a fuzzy neural network based on non-Euclidean distance clustering is proposed in which the input space is partitioned into many local regions by the fuzzy clustering based on non-Euclidean distance so that the computation complexity is decreased, and fuzzy rule number is determined by validity function based on both the separation and the compactness among clusterings. Then, the premise parameters and consequent parameters are trained by hybrid learning algorithm. The parameters identification is realized; meanwhile the convergence condition of consequent parameters is obtained by Lyapunov function. Finally, the proposed method is applied to build the quality index model in slashing process in which the experimental data come from the actual slashing process. The experiment results show that the proposed fuzzy neural network for quality index model has lower computation complexity and faster convergence time, comparing with GP-FNN, BPNN, and RBFNN.
A Neural Networks-Based Hybrid Routing Protocol for Wireless Mesh Networks
Directory of Open Access Journals (Sweden)
Nenad Kojić
2012-06-01
Full Text Available The networking infrastructure of wireless mesh networks (WMNs is decentralized and relatively simple, but they can display reliable functioning performance while having good redundancy. WMNs provide Internet access for fixed and mobile wireless devices. Both in urban and rural areas they provide users with high-bandwidth networks over a specific coverage area. The main problems affecting these networks are changes in network topology and link quality. In order to provide regular functioning, the routing protocol has the main influence in WMN implementations. In this paper we suggest a new routing protocol for WMN, based on good results of a proactive and reactive routing protocol, and for that reason it can be classified as a hybrid routing protocol. The proposed solution should avoid flooding and creating the new routing metric. We suggest the use of artificial logic—i.e., neural networks (NNs. This protocol is based on mobile agent technologies controlled by a Hopfield neural network. In addition to this, our new routing metric is based on multicriteria optimization in order to minimize delay and blocking probability (rejected packets or their retransmission. The routing protocol observes real network parameters and real network environments. As a result of artificial logic intelligence, the proposed routing protocol should maximize usage of network resources and optimize network performance.
Monitoring the Freshness of Moroccan Sardines with a Neural-Network Based Electronic Nose
Directory of Open Access Journals (Sweden)
Benachir Bouchikhi
2006-10-01
Full Text Available An electronic nose was developed and used as a rapid technique to classify thefreshness of sardine samples according to the number of days spent under cold storage (4 Ã‚Â±1Ã‚Â°C, in air. The volatile compounds present in the headspace of weighted sardine sampleswere introduced into a sensor chamber and the response signals of the sensors wererecorded as a function of time. Commercially available gas sensors based on metal oxidesemiconductors were used and both static and dynamic features from the sensorconductance response were input to the pattern recognition engine. Data analysis wasperformed by three different pattern recognition methods such as probabilistic neuralnetworks (PNN, fuzzy ARTMAP neural networks (FANN and support vector machines(SVM. The objective of this study was to find, among these three pattern recognitionmethods, the most suitable one for accurately identifying the days of cold storage undergoneby sardine samples. The results show that the electronic nose can monitor the freshness ofsardine samples stored at 4Ã‚Â°C, and that the best classification and prediction are obtainedwith SVM neural network. The SVM approach shows improved classificationperformances, reducing the amount of misclassified samples down to 3.75 %.
Neural Network Based Response Prediction of rTMS in Major Depressive Disorder Using QEEG Cordance.
Erguzel, Turker Tekin; Ozekes, Serhat; Gultekin, Selahattin; Tarhan, Nevzat; Hizli Sayar, Gokben; Bayram, Ali
2015-01-01
The combination of repetitive transcranial magnetic stimulation (rTMS), a non-pharmacological form of therapy for treating major depressive disorder (MDD), and electroencephalogram (EEG) is a valuable tool for investigating the functional connectivity in the brain. This study aims to explore whether pre-treating frontal quantitative EEG (QEEG) cordance is associated with response to rTMS treatment among MDD patients by using an artificial intelligence approach, artificial neural network (ANN). The artificial neural network using pre-treatment cordance of frontal QEEG classification was carried out to identify responder or non-responder to rTMS treatment among 55 MDD subjects. The classification performance was evaluated using k-fold cross-validation. The ANN classification identified responders to rTMS treatment with a sensitivity of 93.33%, and its overall accuracy reached to 89.09%. Area under Receiver Operating Characteristic (ROC) curve (AUC) value for responder detection using 6, 8 and 10 fold cross validation were 0.917, 0.823 and 0.894 respectively. Potential utility of ANN approach method can be used as a clinical tool in administering rTMS therapy to a targeted group of subjects suffering from MDD. This methodology is more potentially useful to the clinician as prediction is possible using EEG data collected before this treatment process is initiated. It is worth using feature selection algorithms to raise the sensitivity and accuracy values.
Artificial Neural Network Based Model for Forecasting of Inflation in India
Directory of Open Access Journals (Sweden)
Gour Sundar Mitra Thakur
2016-03-01
Full Text Available Inflation can be attributed to both microeconomic and macroeconomic factors which influence the stability of the economy of any nation. With the raising of recession at the end of the year 2008, world communities started paying much contemplation on inflation and put enormous hard work to predict it accurately. Prediction of inflation is not a simple task. Moreover, the behavior of inflation is so complex and uncertain that both economists and statisticians have been striving to model and forecast inflation in an accurate way. As a result, many researchers have proposed inflation forecasting models based on different methods; however the accuracy is always being a major constraint. In this paper, we have analyzed the historical monthly economic data of India between January 2000 and December 2012 and constructed an inflation forecasting model based on feed forward back propagation neural network. Initially some critical factors that can considerably influence the inflation of India have been identified, then an efficient artificial neural network (ANN model has been proposed to forecast the inflation. Accuracy of the model is proved to be satisfactory when compared with the forecasting of some well-known agencies.
A neural networks-based hybrid routing protocol for wireless mesh networks.
Kojić, Nenad; Reljin, Irini; Reljin, Branimir
2012-01-01
The networking infrastructure of wireless mesh networks (WMNs) is decentralized and relatively simple, but they can display reliable functioning performance while having good redundancy. WMNs provide Internet access for fixed and mobile wireless devices. Both in urban and rural areas they provide users with high-bandwidth networks over a specific coverage area. The main problems affecting these networks are changes in network topology and link quality. In order to provide regular functioning, the routing protocol has the main influence in WMN implementations. In this paper we suggest a new routing protocol for WMN, based on good results of a proactive and reactive routing protocol, and for that reason it can be classified as a hybrid routing protocol. The proposed solution should avoid flooding and creating the new routing metric. We suggest the use of artificial logic-i.e., neural networks (NNs). This protocol is based on mobile agent technologies controlled by a Hopfield neural network. In addition to this, our new routing metric is based on multicriteria optimization in order to minimize delay and blocking probability (rejected packets or their retransmission). The routing protocol observes real network parameters and real network environments. As a result of artificial logic intelligence, the proposed routing protocol should maximize usage of network resources and optimize network performance.
[Study on a back propogation neural network-based predictive model for prevalence of birth defect].
Wang, Wei; Xu, Wei; Zheng, Ya-jun; Zhou, Bao-sen
2007-05-01
To evaluate the value of a back propogation (BP) network on prediction of birth defect and to give clues on its prevention. Data of birth defect in Shenyang from 1995 to 2005 were used as a training set to predict the prevalence rate of birth defect. Neural network tools box of Software MATLAB 6.5 was used to train and simulate BP Artificial Neural Network. When using data of the year 1995-2003 to predict the prevalence rate of birth defect in 2004-2005, the results showed that: the fitting average error of prevalence rate was 1.34%, RNL was 0.9874, and the prediction of average error was 1.78%. Using data of the year 1995-2005 to predict the prevalence rate of birth defect in 2006-2007, the results showed that: the fitting average error was 0.33%, RNL was 0.9954, the prevalence rates of birth defect in 2006-2007 were 11.00% and 11.29%. Compared to the conventional statistics method, BP not only showed better prediction precision, but had no limit to the type or distribution of relevant data, thus providing a powerful method in epidemiological prediction.
Ni, Jianjun; Wu, Liuying; Shi, Pengfei; Yang, Simon X
2017-01-01
Real-time path planning for autonomous underwater vehicle (AUV) is a very difficult and challenging task. Bioinspired neural network (BINN) has been used to deal with this problem for its many distinct advantages: that is, no learning process is needed and realization is also easy. However, there are some shortcomings when BINN is applied to AUV path planning in a three-dimensional (3D) unknown environment, including complex computing problem when the environment is very large and repeated path problem when the size of obstacles is bigger than the detection range of sensors. To deal with these problems, an improved dynamic BINN is proposed in this paper. In this proposed method, the AUV is regarded as the core of the BINN and the size of the BINN is based on the detection range of sensors. Then the BINN will move with the AUV and the computing could be reduced. A virtual target is proposed in the path planning method to ensure that the AUV can move to the real target effectively and avoid big-size obstacles automatically. Furthermore, a target attractor concept is introduced to improve the computing efficiency of neural activities. Finally, some experiments are conducted under various 3D underwater environments. The experimental results show that the proposed BINN based method can deal with the real-time path planning problem for AUV efficiently.
Directory of Open Access Journals (Sweden)
Si-Yao Fu
2012-01-01
Full Text Available In this paper, we present a quantitative, highly structured cortex-simulated model, which can be simply described as feedforward, hierarchical simulation of ventral stream of visual cortex using biologically plausible, computationally convenient spiking neural network system. The motivation comes directly from recent pioneering works on detailed functional decomposition analysis of the feedforward pathway of the ventral stream of visual cortex and developments on artificial spiking neural networks (SNNs. By combining the logical structure of the cortical hierarchy and computing power of the spiking neuron model, a practical framework has been presented. As a proof of principle, we demonstrate our system on several facial expression recognition tasks. The proposed cortical-like feedforward hierarchy framework has the merit of capability of dealing with complicated pattern recognition problems, suggesting that, by combining the cognitive models with modern neurocomputational approaches, the neurosystematic approach to the study of cortex-like mechanism has the potential to extend our knowledge of brain mechanisms underlying the cognitive analysis and to advance theoretical models of how we recognize face or, more specifically, perceive other people’s facial expression in a rich, dynamic, and complex environment, providing a new starting point for improved models of visual cortex-like mechanism.
Directory of Open Access Journals (Sweden)
Farahnaz SADOUGHI
2014-03-01
Full Text Available Breast cancer is the most commonly diagnosed cancer and the most common cause of death in women all over the world. Use of computer technology supporting breast cancer diagnosing is now widespread and pervasive across a broad range of medical areas. Early diagnosis of this disease can greatly enhance the chances of long-term survival of breast cancer victims. Artificial Neural Networks (ANN as mainly method play important role in early diagnoses breast cancer. This paper studies Levenberg Marquardet Backpropagation (LMBP neural network and Levenberg Marquardet Backpropagation based Particle Swarm Optimization(LMBP-PSO for the diagnosis of breast cancer. The obtained results show that LMBP and LMBP based PSO system provides higher classification efficiency. But LMBP based PSO needs minimum training and testing time. It helps in developing Medical Decision System (MDS for breast cancer diagnosing. It can also be used as secondary observer in clinical decision making.
Directory of Open Access Journals (Sweden)
Qiang Gao
2016-01-01
Full Text Available To satisfy the lightweight requirements of large pipe weapons, a novel electrohydraulic servo (EHS system where the hydraulic cylinder possesses three cavities is developed and investigated in the present study. In the EHS system, the balancing cavity of the EHS is especially designed for active compensation for the unbalancing force of the system, whereas the two driving cavities are employed for positioning and disturbance rejection of the large pipe. Aiming at simultaneously balancing and positioning of the EHS system, a novel neural network based active disturbance rejection control (NNADRC strategy is developed. In the NNADRC, the radial basis function (RBF neural network is employed for online updating of parameters of the extended state observer (ESO. Thereby, the nonlinear behavior and external disturbance of the system can be accurately estimated and compensated in real time. The efficiency and superiority of the system are critically investigated by conducting numerical simulations, showing that much higher steady accuracy as well as system robustness is achieved when comparing with conventional ADRC control system. It indicates that the NNADRC is a very promising technique for achieving fast, stable, smooth, and accurate control of the novel EHS system.
Energy Technology Data Exchange (ETDEWEB)
Coppini, G. [Consiglio Nazionale delle Ricerche, Pisa (Italy). Ist. di Fisiologia Patologica; Valli, G. [Florence Univ., Florence (Italy). Dipt. di Ingegneria Elettronica; Falchini, M.; Stecco, A.; Bindi, A.; Carmignani, L. [Florence Univ., Florence (Italy). Dipt. di Fisiopatologia Clinica, Sezione di Radiodiagnostica
1999-10-01
In this report are investigated the capabilities of an artificial neural network-based Computer-Aided Diagnosis (CAD) system in improving early detection of pulmonary nodules on chest radiographs. [Italian] Valutazione di un sistema di riconoscimento automatico basato sulla tecnologia delle reti neruali artificiali per migliorare le possibilita' di rivelazione precoce dei noduli polmonari sul radiogramma toracico.
DEFF Research Database (Denmark)
May, Tobias
2018-01-01
This study presents an algorithm for binaural speech dereverberation based on the supervised learning of short-term binaural cues. The proposed system combined a delay-and-sum beamformer (DSB) with a neural network-based post-filter that attenuated reverberant components in individual time...
Neural network-based optimal adaptive output feedback control of a helicopter UAV.
Nodland, David; Zargarzadeh, Hassan; Jagannathan, Sarangapani
2013-07-01
Helicopter unmanned aerial vehicles (UAVs) are widely used for both military and civilian operations. Because the helicopter UAVs are underactuated nonlinear mechanical systems, high-performance controller design for them presents a challenge. This paper introduces an optimal controller design via an output feedback for trajectory tracking of a helicopter UAV, using a neural network (NN). The output-feedback control system utilizes the backstepping methodology, employing kinematic and dynamic controllers and an NN observer. The online approximator-based dynamic controller learns the infinite-horizon Hamilton-Jacobi-Bellman equation in continuous time and calculates the corresponding optimal control input by minimizing a cost function, forward-in-time, without using the value and policy iterations. Optimal tracking is accomplished by using a single NN utilized for the cost function approximation. The overall closed-loop system stability is demonstrated using Lyapunov analysis. Finally, simulation results are provided to demonstrate the effectiveness of the proposed control design for trajectory tracking.
Cellular Neural Network-Based Methods for Distributed Network Intrusion Detection
Directory of Open Access Journals (Sweden)
Kang Xie
2015-01-01
Full Text Available According to the problems of current distributed architecture intrusion detection systems (DIDS, a new online distributed intrusion detection model based on cellular neural network (CNN was proposed, in which discrete-time CNN (DTCNN was used as weak classifier in each local node and state-controlled CNN (SCCNN was used as global detection method, respectively. We further proposed a new method for design template parameters of SCCNN via solving Linear Matrix Inequality. Experimental results based on KDD CUP 99 dataset show its feasibility and effectiveness. Emerging evidence has indicated that this new approach is affordable to parallelism and analog very large scale integration (VLSI implementation which allows the distributed intrusion detection to be performed better.
Ding, Zhixia; Shen, Yi
2016-04-01
This paper investigates global projective synchronization of nonidentical fractional-order neural networks (FNNs) based on sliding mode control technique. We firstly construct a fractional-order integral sliding surface. Then, according to the sliding mode control theory, we design a sliding mode controller to guarantee the occurrence of the sliding motion. Based on fractional Lyapunov direct methods, system trajectories are driven to the proposed sliding surface and remain on it evermore, and some novel criteria are obtained to realize global projective synchronization of nonidentical FNNs. As the special cases, some sufficient conditions are given to ensure projective synchronization of identical FNNs, complete synchronization of nonidentical FNNs and anti-synchronization of nonidentical FNNs. Finally, one numerical example is given to demonstrate the effectiveness of the obtained results. Copyright © 2016 Elsevier Ltd. All rights reserved.
Neural Network-Based Receiver in Band-Limited Communication System with MPPSK Modulation
Directory of Open Access Journals (Sweden)
Wang Zixin
2018-01-01
Full Text Available As a type of the spectrally efficient modulation, the m-ary phase position shift keying (MPPSK has been considered to meet the increasing spectrum requirement in the future wireless system. To limit the signal bandwidth and cancel the out-band interference the band-pass filters are used, which introduce the waveform distortion and inter-symbol interference (ISI. Therefore, a single hidden-layer neural network (NN-based receiver is proposed to jointly equalize and demodulate the received signal. The impulse response of the system is static and the network parameters can be obtained after off-line training. The number of the hidden nodes is also determined through simulations. Simulation results show that the NN-based receiver works well in the communication system with different allocated bandwidths. By observing the modified confusion matrix, the false symbol decision is relevant to modulation index, waveform distortions and the ISI.
Performance Evaluation of Neural Network Based Pulse-Echo Weld Defect Classifiers
Seyedtabaii, S.
2012-10-01
Pulse-echo ultrasonic signal is used to detect weld defects with high probability. However, utilizing echo signal for defects classification is another issue that has attracted attention of many researchers who have devised algorithms and tested them against their own databases. In this paper, a study is conducted to score the performance of various algorithms against a single echo signal database. Algorithms tested the use of Wavelet Transform (WT), Fast Fourier Transform (FFT) and time domain echo signal features and employed several NN’s architectures such as Multi-Layer Perceptron Neural Network (MLP), Self Organizing Map (SOM) and others known to be good classifiers. The average performance of all can be viewed fair (90%) while some algorithms render success rate of about 94%. It seems that acquiring higher success rates out of a single fixed angle probe pulseecho set up needs new arrangements of data collection, which is under investigation.
ARTIFICIAL NEURAL NETWORK BASED ULTRASONIC SENSOR SYSTEM FOR DETECTION OF ADULTERATION IN EDIBLE OIL
Directory of Open Access Journals (Sweden)
TONY GEORGE
2017-06-01
Full Text Available This paper presents the design, development and experimental validation of an ultrasonic sensor system for the detection of adulteration in edible oil. Variation of ultrasonic wave propagation characteristics like attenuation coefficient, reflection coefficient and velocity of propagation in pure and adulterated oil were used for developing the algorithm to detect the adulteration. Measurement cell was designed for operating ultrasonic transducer at 1 MHz using COMSOL 4.4. Artificial Neural Network (ANN based algorithm was also developed for improving the efficiency of the sensor system. It is found that this system can detect adulteration with an accuracy of 99.53% for sunflower oil added in pure coconut oil, whereas 98.82% for palm oil added in pure coconut oil.
Managing Emergencies Optimally Using a Random Neural Network-Based Algorithm
Directory of Open Access Journals (Sweden)
Qing Han
2013-10-01
Full Text Available Emergency rescues require that first responders provide support to evacuate injured and other civilians who are obstructed by the hazards. In this case, the emergency personnel can take actions strategically in order to rescue people maximally, efficiently and quickly. The paper studies the effectiveness of a random neural network (RNN-based task assignment algorithm involving optimally matching emergency personnel and injured civilians, so that the emergency personnel can aid trapped people to move towards evacuation exits in real-time. The evaluations are run on a decision support evacuation system using the Distributed Building Evacuation Simulator (DBES multi-agent platform in various emergency scenarios. The simulation results indicate that the RNN-based task assignment algorithm provides a near-optimal solution to resource allocation problems, which avoids resource wastage and improves the efficiency of the emergency rescue process.
Convolutional Neural Network-Based Finger-Vein Recognition Using NIR Image Sensors.
Hong, Hyung Gil; Lee, Min Beom; Park, Kang Ryoung
2017-06-06
Conventional finger-vein recognition systems perform recognition based on the finger-vein lines extracted from the input images or image enhancement, and texture feature extraction from the finger-vein images. In these cases, however, the inaccurate detection of finger-vein lines lowers the recognition accuracy. In the case of texture feature extraction, the developer must experimentally decide on a form of the optimal filter for extraction considering the characteristics of the image database. To address this problem, this research proposes a finger-vein recognition method that is robust to various database types and environmental changes based on the convolutional neural network (CNN). In the experiments using the two finger-vein databases constructed in this research and the SDUMLA-HMT finger-vein database, which is an open database, the method proposed in this research showed a better performance compared to the conventional methods.
Directory of Open Access Journals (Sweden)
Jian Yang
2016-09-01
Full Text Available In this article, a tree search algorithm is proposed to find the near optimal conflict avoidance solutions for unmanned aerial vehicles. In the dynamic environment, the unmodeled elements, such as wind, would make UAVs deviate from nominal traces. It brings about difficulties for conflict detection and resolution. The back propagation neural networks are utilized to approximate the unmodeled dynamics of the environment. To satisfy the online planning requirement, the search length of the tree search algorithm would be limited. Therefore, the algorithm may not be able to reach the goal states in search process. The midterm reward function for assessing each node is devised, with consideration given to two factors, namely, the safe separation requirement and the mission of each unmanned aerial vehicle. The simulation examples and the comparisons with previous approaches are provided to illustrate the smooth and convincing behaviours of the proposed algorithm.
A Neural-Network-Based Semi-Automated Geospatial Classification Tool
Hale, R. G.; Herzfeld, U. C.
2014-12-01
North America's largest glacier system, the Bering Bagley Glacier System (BBGS) in Alaska, surged in 2011-2013, as shown by rapid mass transfer, elevation change, and heavy crevassing. Little is known about the physics controlling surge glaciers' semi-cyclic patterns; therefore, it is crucial to collect and analyze as much data as possible so that predictive models can be made. In addition, physical signs frozen in ice in the form of crevasses may help serve as a warning for future surges. The BBGS surge provided an opportunity to develop an automated classification tool for crevasse classification based on imagery collected from small aircraft. The classification allows one to link image classification to geophysical processes associated with ice deformation. The tool uses an approach that employs geostatistical functions and a feed-forward perceptron with error back-propagation. The connectionist-geostatistical approach uses directional experimental (discrete) variograms to parameterize images into a form that the Neural Network (NN) can recognize. In an application to preform analysis on airborne video graphic data from the surge of the BBGS, an NN was able to distinguish 18 different crevasse classes with 95 percent or higher accuracy, for over 3,000 images. Recognizing that each surge wave results in different crevasse types and that environmental conditions affect the appearance in imagery, we designed the tool's semi-automated pre-training algorithm to be adaptable. The tool can be optimized to specific settings and variables of image analysis: (airborne and satellite imagery, different camera types, observation altitude, number and types of classes, and resolution). The generalization of the classification tool brings three important advantages: (1) multiple types of problems in geophysics can be studied, (2) the training process is sufficiently formalized to allow non-experts in neural nets to perform the training process, and (3) the time required to
Deep architecture neural network-based real-time image processing for image-guided radiotherapy.
Mori, Shinichiro
2017-08-01
To develop real-time image processing for image-guided radiotherapy, we evaluated several neural network models for use with different imaging modalities, including X-ray fluoroscopic image denoising. Setup images of prostate cancer patients were acquired with two oblique X-ray fluoroscopic units. Two types of residual network were designed: a convolutional autoencoder (rCAE) and a convolutional neural network (rCNN). We changed the convolutional kernel size and number of convolutional layers for both networks, and the number of pooling and upsampling layers for rCAE. The ground-truth image was applied to the contrast-limited adaptive histogram equalization (CLAHE) method of image processing. Network models were trained to keep the quality of the output image close to that of the ground-truth image from the input image without image processing. For image denoising evaluation, noisy input images were used for the training. More than 6 convolutional layers with convolutional kernels >5×5 improved image quality. However, this did not allow real-time imaging. After applying a pair of pooling and upsampling layers to both networks, rCAEs with >3 convolutions each and rCNNs with >12 convolutions with a pair of pooling and upsampling layers achieved real-time processing at 30 frames per second (fps) with acceptable image quality. Use of our suggested network achieved real-time image processing for contrast enhancement and image denoising by the use of a conventional modern personal computer. Copyright © 2017 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved.
Estimating Conditional Distributions by Neural Networks
DEFF Research Database (Denmark)
Kulczycki, P.; Schiøler, Henrik
1998-01-01
Neural Networks for estimating conditionaldistributions and their associated quantiles are investigated in this paper. A basic network structure is developed on the basis of kernel estimation theory, and consistency property is considered from a mild set of assumptions. A number of applications...
Neural Network for Estimating Conditional Distribution
DEFF Research Database (Denmark)
Schiøler, Henrik; Kulczycki, P.
Neural networks for estimating conditional distributions and their associated quantiles are investigated in this paper. A basic network structure is developed on the basis of kernel estimation theory, and consistency is proved from a mild set of assumptions. A number of applications within...... statistcs, decision theory and signal processing are suggested, and a numerical example illustrating the capabilities of the elaborated network is given...
Yan, Qin; Zhong, Yanfei
2008-12-01
The radial basis function (RBF) neural network is a powerful method for remote sensing image classification. It has a simple architecture and the learning algorithm corresponds to the solution of a linear regression problem, resulting in a fast training process. The main drawback of this strategy is the requirement of an efficient algorithm to determine the number, position, and dispersion of the RBF. Traditional methods to determine the centers are: randomly choose input vectors from the training data set; vectors obtained from unsupervised clustering algorithms, such as k-means, applied to the input data. These conduce that traditional RBF neural network is sensitive to the center initialization. In this paper, the artificial immune network (aiNet) model, a new computational intelligence based on artificial immune networks (AIN), is applied to obtain appropriate centers for remote sensing image classification. In the aiNet-RBF algorihtm, each input pattern corresonds to an antigenic stimulus, while each RBF candidate center is considered to be an element, or cell, of the immune network model. The steps are as follows: A set of candidate centers is initialized at random, where the initial number of candidates and their positions is not crucial to the performance. Then, the clonal selection principle will control which candidates will be selected and how they will be upadated. Note that the clonal selection principle will be responsible for how the centers will represent the training data set. Finally, the immune network will identify and eliminate or suppress self-recognizing individuals to control the number of candidate centers. After the above learning phase, the aiNet network centers represent internal images of the inuput patterns presented to it. The algorithm output is taken to be the matrix of memory cells' coordinates that represent the final centers to be adopted by the RBF network. The stopping criterion of the proposed algorithm is given by a pre
Gesture Recognition Using Neural Networks Based on HW/SW Cosimulation Platform
Directory of Open Access Journals (Sweden)
Priyanka Mekala
2013-01-01
Full Text Available Hardware/software (HW/SW cosimulation integrates software simulation and hardware simulation simultaneously. Usually, HW/SW co-simulation platform is used to ease debugging and verification for very large-scale integration (VLSI design. To accelerate the computation of the gesture recognition technique, an HW/SW implementation using field programmable gate array (FPGA technology is presented in this paper. The major contributions of this work are: (1 a novel design of memory controller in the Verilog Hardware Description Language (Verilog HDL to reduce memory consumption and load on the processor. (2 The testing part of the neural network algorithm is being hardwired to improve the speed and performance. The American Sign Language gesture recognition is chosen to verify the performance of the approach. Several experiments were carried out on four databases of the gestures (alphabet signs A to Z. (3 The major benefit of this design is that it takes only few milliseconds to recognize the hand gesture which makes it computationally more efficient.
Visual Servoing for an Autonomous Hexarotor Using a Neural Network Based PID Controller.
Lopez-Franco, Carlos; Gomez-Avila, Javier; Alanis, Alma Y; Arana-Daniel, Nancy; Villaseñor, Carlos
2017-08-12
In recent years, unmanned aerial vehicles (UAVs) have gained significant attention. However, we face two major drawbacks when working with UAVs: high nonlinearities and unknown position in 3D space since it is not provided with on-board sensors that can measure its position with respect to a global coordinate system. In this paper, we present a real-time implementation of a servo control, integrating vision sensors, with a neural proportional integral derivative (PID), in order to develop an hexarotor image based visual servo control (IBVS) that knows the position of the robot by using a velocity vector as a reference to control the hexarotor position. This integration requires a tight coordination between control algorithms, models of the system to be controlled, sensors, hardware and software platforms and well-defined interfaces, to allow the real-time implementation, as well as the design of different processing stages with their respective communication architecture. All of these issues and others provoke the idea that real-time implementations can be considered as a difficult task. For the purpose of showing the effectiveness of the sensor integration and control algorithm to address these issues on a high nonlinear system with noisy sensors as cameras, experiments were performed on the Asctec Firefly on-board computer, including both simulation and experimenta results.
Neural network-based motion control of an underactuated wheeled inverted pendulum model.
Yang, Chenguang; Li, Zhijun; Cui, Rongxin; Xu, Bugong
2014-11-01
In this paper, automatic motion control is investigated for one of wheeled inverted pendulum (WIP) models, which have been widely applied for modeling of a large range of two wheeled modern vehicles. First, the underactuated WIP model is decomposed into a fully actuated second order subsystem Σa consisting of planar movement of vehicle forward and yaw angular motions, and a nonactuated first order subsystem Σb of pendulum motion. Due to the unknown dynamics of subsystem Σa and the universal approximation ability of neural network (NN), an adaptive NN scheme has been employed for motion control of subsystem Σa . The model reference approach has been used whereas the reference model is optimized by the finite time linear quadratic regulation technique. The pendulum motion in the passive subsystem Σb is indirectly controlled using the dynamic coupling with planar forward motion of subsystem Σa , such that satisfactory tracking of a set pendulum tilt angle can be guaranteed. Rigours theoretic analysis has been established, and simulation studies have been performed to demonstrate the developed method.
Mandal, Sudip; Saha, Goutam; Pal, Rajat Kumar
2017-08-01
Correct inference of genetic regulations inside a cell from the biological database like time series microarray data is one of the greatest challenges in post genomic era for biologists and researchers. Recurrent Neural Network (RNN) is one of the most popular and simple approach to model the dynamics as well as to infer correct dependencies among genes. Inspired by the behavior of social elephants, we propose a new metaheuristic namely Elephant Swarm Water Search Algorithm (ESWSA) to infer Gene Regulatory Network (GRN). This algorithm is mainly based on the water search strategy of intelligent and social elephants during drought, utilizing the different types of communication techniques. Initially, the algorithm is tested against benchmark small and medium scale artificial genetic networks without and with presence of different noise levels and the efficiency was observed in term of parametric error, minimum fitness value, execution time, accuracy of prediction of true regulation, etc. Next, the proposed algorithm is tested against the real time gene expression data of Escherichia Coli SOS Network and results were also compared with others state of the art optimization methods. The experimental results suggest that ESWSA is very efficient for GRN inference problem and performs better than other methods in many ways.
Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification
Directory of Open Access Journals (Sweden)
Srdjan Sladojevic
2016-01-01
Full Text Available The latest generation of convolutional neural networks (CNNs has achieved impressive results in the field of image classification. This paper is concerned with a new approach to the development of plant disease recognition model, based on leaf image classification, by the use of deep convolutional networks. Novel way of training and the methodology used facilitate a quick and easy system implementation in practice. The developed model is able to recognize 13 different types of plant diseases out of healthy leaves, with the ability to distinguish plant leaves from their surroundings. According to our knowledge, this method for plant disease recognition has been proposed for the first time. All essential steps required for implementing this disease recognition model are fully described throughout the paper, starting from gathering images in order to create a database, assessed by agricultural experts. Caffe, a deep learning framework developed by Berkley Vision and Learning Centre, was used to perform the deep CNN training. The experimental results on the developed model achieved precision between 91% and 98%, for separate class tests, on average 96.3%.
Zargarzadeh, H.; Nodland, David; Thotla, V.; Jagannathan, S.; Agarwal, S.
2012-06-01
Unmanned Aerial Vehicles (UAVs) are versatile aircraft with many applications, including the potential for use to detect unintended electromagnetic emissions from electronic devices. A particular area of recent interest has been helicopter unmanned aerial vehicles. Because of the nature of these helicopters' dynamics, high-performance controller design for them presents a challenge. This paper introduces an optimal controller design via output feedback control for trajectory tracking of a helicopter UAV using a neural network (NN). The output-feedback control system utilizes the backstepping methodology, employing kinematic, virtual, and dynamic controllers and an observer. Optimal tracking is accomplished with a single NN utilized for cost function approximation. The controller positions the helicopter, which is equipped with an antenna, such that the antenna can detect unintended emissions. The overall closed-loop system stability with the proposed controller is demonstrated by using Lyapunov analysis. Finally, results are provided to demonstrate the effectiveness of the proposed control design for positioning the helicopter for unintended emissions detection.
CERN. Geneva
2017-01-01
Parameter tuning is an important task of storage performance optimization. Current practice usually involves numerous tweak-benchmark cycles that are slow and costly. To address this issue, we developed CAPES, a model-less deep reinforcement learning-based unsupervised parameter tuning system driven by a deep neural network (DNN). It is designed to nd the optimal values of tunable parameters in computer systems, from a simple client-server system to a large data center, where human tuning can be costly and often cannot achieve optimal performance. CAPES takes periodic measurements of a target computer system’s state, and trains a DNN which uses Q-learning to suggest changes to the system’s current parameter values. CAPES is minimally intrusive, and can be deployed into a production system to collect training data and suggest tuning actions during the system’s daily operation. Evaluation of a prototype on a Lustre system demonstrates an increase in I/O throughput up to 45% at saturation point. About the...
Response surface and neural network based predictive models of cutting temperature in hard turning
Directory of Open Access Journals (Sweden)
Mozammel Mia
2016-11-01
Full Text Available The present study aimed to develop the predictive models of average tool-workpiece interface temperature in hard turning of AISI 1060 steels by coated carbide insert. The Response Surface Methodology (RSM and Artificial Neural Network (ANN were employed to predict the temperature in respect of cutting speed, feed rate and material hardness. The number and orientation of the experimental trials, conducted in both dry and high pressure coolant (HPC environments, were planned using full factorial design. The temperature was measured by using the tool-work thermocouple. In RSM model, two quadratic equations of temperature were derived from experimental data. The analysis of variance (ANOVA and mean absolute percentage error (MAPE were performed to suffice the adequacy of the models. In ANN model, 80% data were used to train and 20% data were employed for testing. Like RSM, herein, the error analysis was also conducted. The accuracy of the RSM and ANN model was found to be ⩾99%. The ANN models exhibit an error of ∼5% MAE for testing data. The regression coefficient was found to be greater than 99.9% for both dry and HPC. Both these models are acceptable, although the ANN model demonstrated a higher accuracy. These models, if employed, are expected to provide a better control of cutting temperature in turning of hardened steel.
Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification
Sladojevic, Srdjan; Arsenovic, Marko; Culibrk, Dubravko; Stefanovic, Darko
2016-01-01
The latest generation of convolutional neural networks (CNNs) has achieved impressive results in the field of image classification. This paper is concerned with a new approach to the development of plant disease recognition model, based on leaf image classification, by the use of deep convolutional networks. Novel way of training and the methodology used facilitate a quick and easy system implementation in practice. The developed model is able to recognize 13 different types of plant diseases out of healthy leaves, with the ability to distinguish plant leaves from their surroundings. According to our knowledge, this method for plant disease recognition has been proposed for the first time. All essential steps required for implementing this disease recognition model are fully described throughout the paper, starting from gathering images in order to create a database, assessed by agricultural experts. Caffe, a deep learning framework developed by Berkley Vision and Learning Centre, was used to perform the deep CNN training. The experimental results on the developed model achieved precision between 91% and 98%, for separate class tests, on average 96.3%. PMID:27418923
Artificial neural networks based controller for glucose monitoring during clamp test.
Directory of Open Access Journals (Sweden)
Merav Catalogna
Full Text Available Insulin resistance (IR is one of the most widespread health problems in modern times. The gold standard for quantification of IR is the hyperinsulinemic-euglycemic glucose clamp technique. During the test, a regulated glucose infusion is delivered intravenously to maintain a constant blood glucose concentration. Current control algorithms for regulating this glucose infusion are based on feedback control. These models require frequent sampling of blood, and can only partly capture the complexity associated with regulation of glucose. Here we present an improved clamp control algorithm which is motivated by the stochastic nature of glucose kinetics, while using the minimal need in blood samples required for evaluation of IR. A glucose pump control algorithm, based on artificial neural networks model was developed. The system was trained with a data base collected from 62 rat model experiments, using a back-propagation Levenberg-Marquardt optimization. Genetic algorithm was used to optimize network topology and learning features. The predictive value of the proposed algorithm during the temporal period of interest was significantly improved relative to a feedback control applied at an equivalent low sampling interval. Robustness to noise analysis demonstrates the applicability of the algorithm in realistic situations.
Visual Servoing for an Autonomous Hexarotor Using a Neural Network Based PID Controller
Lopez-Franco, Carlos; Alanis, Alma Y.; Arana-Daniel, Nancy; Villaseñor, Carlos
2017-01-01
In recent years, unmanned aerial vehicles (UAVs) have gained significant attention. However, we face two major drawbacks when working with UAVs: high nonlinearities and unknown position in 3D space since it is not provided with on-board sensors that can measure its position with respect to a global coordinate system. In this paper, we present a real-time implementation of a servo control, integrating vision sensors, with a neural proportional integral derivative (PID), in order to develop an hexarotor image based visual servo control (IBVS) that knows the position of the robot by using a velocity vector as a reference to control the hexarotor position. This integration requires a tight coordination between control algorithms, models of the system to be controlled, sensors, hardware and software platforms and well-defined interfaces, to allow the real-time implementation, as well as the design of different processing stages with their respective communication architecture. All of these issues and others provoke the idea that real-time implementations can be considered as a difficult task. For the purpose of showing the effectiveness of the sensor integration and control algorithm to address these issues on a high nonlinear system with noisy sensors as cameras, experiments were performed on the Asctec Firefly on-board computer, including both simulation and experimenta results. PMID:28805689
Development of Artificial Neural-Network-Based Models for the Simulation of Spring Discharge
Directory of Open Access Journals (Sweden)
M. Mohan Raju
2011-01-01
Full Text Available The present study demonstrates the application of artificial neural networks (ANNs in predicting the weekly spring discharge. The study was based on the weekly spring discharge from a spring located near Ranichauri in Tehri Garhwal district of Uttarakhand, India. Five models were developed for predicting the spring discharge based on a weekly interval using rainfall, evaporation, temperature with a specified lag time. All models were developed both with one and two hidden layers. Each model was developed with many trials by selecting different network architectures and different number of hidden neurons; finally a best predicting model presented against each developed model. The models were trained with three different algorithms, that is, quick-propagation algorithm, batch backpropagation algorithm, and Levenberg-Marquardt algorithm using weekly data from 1999 to 2005. A best model for the simulation was selected from the three presented algorithms using the statistical criteria such as correlation coefficient (, determination coefficient, or Nash Sutcliff's efficiency (DC. Finally, optimized number of neurons were considered for the best model. Training and testing results revealed that the models were predicting the weekly spring discharge satisfactorily. Based on these criteria, ANN-based model results in better agreement for the computation of spring discharge. LMR models were also developed in the study, and they also gave good results, but, when compared with the ANN methodology, ANN resulted in better optimized values.
A NEURAL NETWORK BASED TRAFFIC-AWARE FORWARDING STRATEGY IN NAMED DATA NETWORKING
Directory of Open Access Journals (Sweden)
Parisa Bazmi
2016-11-01
Full Text Available Named Data Networking (NDN is a new Internet architecture which has been proposed to eliminate TCP/IP Internet architecture restrictions. This architecture is abstracting away the notion of host and working based on naming datagrams. However, one of the major challenges of NDN is supporting QoS-aware forwarding strategy so as to forward Interest packets intelligently over multiple paths based on the current network condition. In this paper, Neural Network (NN Based Traffic-aware Forwarding strategy (NNTF is introduced in order to determine an optimal path for Interest forwarding. NN is embedded in NDN routers to select next hop dynamically based on the path overload probability achieved from the NN. This solution is characterized by load balancing and QoS-awareness via monitoring the available path and forwarding data on the traffic-aware shortest path. The performance of NNTF is evaluated using ndnSIM which shows the efficiency of this scheme in terms of network QoS improvementof17.5% and 72% reduction in network delay and packet drop respectively.
Dhawan, Atam P.; Chitre, Yateen S.; Moskowitz, Myron
1993-07-01
Mammography associated with clinical breast examination and self-breast examination is the only effective and viable method for mass breast screening. It is however, difficult to distinguish between benign and malignant microcalcifications associated with breast cancer. Most of the techniques used in the computerized analysis of mammographic microcalcifications segment the digitized gray-level image into regions representing microcalcifications. We present a second-order gray-level histogram based feature extraction approach to extract microcalcification features. These features, called image structure features, are computed from the second-order gray-level histogram statistics, and do not require segmentation of the original image into binary regions. Several image structure features were computed for 100 cases of `difficult to diagnose' microcalcification cases with known biopsy results. These features were analyzed in a correlation study which provided a set of five best image structure features. A feedforward backpropagation neural network was used to classify mammographic microcalcifications using the image structure features. The network was trained on 10 cases of mammographic microcalcifications and tested on additional 85 `difficult-to-diagnose' microcalcifications cases using the selected image structure features. The trained network yielded good results for classification of `difficult-to- diagnose' microcalcifications into benign and malignant categories.
Response surface and neural network based predictive models of cutting temperature in hard turning.
Mia, Mozammel; Dhar, Nikhil R
2016-11-01
The present study aimed to develop the predictive models of average tool-workpiece interface temperature in hard turning of AISI 1060 steels by coated carbide insert. The Response Surface Methodology (RSM) and Artificial Neural Network (ANN) were employed to predict the temperature in respect of cutting speed, feed rate and material hardness. The number and orientation of the experimental trials, conducted in both dry and high pressure coolant (HPC) environments, were planned using full factorial design. The temperature was measured by using the tool-work thermocouple. In RSM model, two quadratic equations of temperature were derived from experimental data. The analysis of variance (ANOVA) and mean absolute percentage error (MAPE) were performed to suffice the adequacy of the models. In ANN model, 80% data were used to train and 20% data were employed for testing. Like RSM, herein, the error analysis was also conducted. The accuracy of the RSM and ANN model was found to be ⩾99%. The ANN models exhibit an error of ∼5% MAE for testing data. The regression coefficient was found to be greater than 99.9% for both dry and HPC. Both these models are acceptable, although the ANN model demonstrated a higher accuracy. These models, if employed, are expected to provide a better control of cutting temperature in turning of hardened steel.
Stratified construction of neural network based interatomic models for multicomponent materials
Hajinazar, Samad; Shao, Junping; Kolmogorov, Aleksey N.
2017-01-01
Recent application of neural networks (NNs) to modeling interatomic interactions has shown the learning machines' encouragingly accurate performance for select elemental and multicomponent systems. In this study we explore the possibility of building a library of NN-based models by introducing a hierarchical NN training. In such a stratified procedure NNs for multicomponent systems are obtained by sequential training from the bottom up: first unaries, then binaries, and so on. Advantages of constructing NN sets with shared parameters include acceleration of the training process and intact description of the constituent systems. We use an automated generation of diverse structure sets for NN training on density functional theory-level reference energies. In the test case of Cu, Pd, Ag, Cu-Pd, Cu-Ag, Pd-Ag, and Cu-Pd-Ag systems, NNs trained in the traditional and stratified fashions are found to have essentially identical accuracy for defect energies, phonon dispersions, formation energies, etc. The models' robustness is further illustrated via unconstrained evolutionary structure searches in which the NN is used for the local optimization of crystal unit cells.
Abstract Computation in Schizophrenia Detection through Artificial Neural Network Based Systems
Directory of Open Access Journals (Sweden)
L. Cardoso
2015-01-01
Full Text Available Schizophrenia stands for a long-lasting state of mental uncertainty that may bring to an end the relation among behavior, thought, and emotion; that is, it may lead to unreliable perception, not suitable actions and feelings, and a sense of mental fragmentation. Indeed, its diagnosis is done over a large period of time; continuos signs of the disturbance persist for at least 6 (six months. Once detected, the psychiatrist diagnosis is made through the clinical interview and a series of psychic tests, addressed mainly to avoid the diagnosis of other mental states or diseases. Undeniably, the main problem with identifying schizophrenia is the difficulty to distinguish its symptoms from those associated to different untidiness or roles. Therefore, this work will focus on the development of a diagnostic support system, in terms of its knowledge representation and reasoning procedures, based on a blended of Logic Programming and Artificial Neural Networks approaches to computing, taking advantage of a novel approach to knowledge representation and reasoning, which aims to solve the problems associated in the handling (i.e., to stand for and reason of defective information.
A pre-trained convolutional neural network based method for thyroid nodule diagnosis.
Ma, Jinlian; Wu, Fa; Zhu, Jiang; Xu, Dong; Kong, Dexing
2017-01-01
In ultrasound images, most thyroid nodules are in heterogeneous appearances with various internal components and also have vague boundaries, so it is difficult for physicians to discriminate malignant thyroid nodules from benign ones. In this study, we propose a hybrid method for thyroid nodule diagnosis, which is a fusion of two pre-trained convolutional neural networks (CNNs) with different convolutional layers and fully-connected layers. Firstly, the two networks pre-trained with ImageNet database are separately trained. Secondly, we fuse feature maps learned by trained convolutional filters, pooling and normalization operations of the two CNNs. Finally, with the fused feature maps, a softmax classifier is used to diagnose thyroid nodules. The proposed method is validated on 15,000 ultrasound images collected from two local hospitals. Experiment results show that the proposed CNN based methods can accurately and effectively diagnose thyroid nodules. In addition, the fusion of the two CNN based models lead to significant performance improvement, with an accuracy of 83.02%±0.72%. These demonstrate the potential clinical applications of this method. Copyright © 2016 Elsevier B.V. All rights reserved.
Seeliger, K; Fritsche, M; Güçlü, U; Schoenmakers, S; Schoffelen, J-M; Bosch, S E; van Gerven, M A J
2017-07-16
Representations learned by deep convolutional neural networks (CNNs) for object recognition are a widely investigated model of the processing hierarchy in the human visual system. Using functional magnetic resonance imaging, CNN representations of visual stimuli have previously been shown to correspond to processing stages in the ventral and dorsal streams of the visual system. Whether this correspondence between models and brain signals also holds for activity acquired at high temporal resolution has been explored less exhaustively. Here, we addressed this question by combining CNN-based encoding models with magnetoencephalography (MEG). Human participants passively viewed 1,000 images of objects while MEG signals were acquired. We modelled their high temporal resolution source-reconstructed cortical activity with CNNs, and observed a feed-forward sweep across the visual hierarchy between 75 and 200 ms after stimulus onset. This spatiotemporal cascade was captured by the network layer representations, where the increasingly abstract stimulus representation in the hierarchical network model was reflected in different parts of the visual cortex, following the visual ventral stream. We further validated the accuracy of our encoding model by decoding stimulus identity in a left-out validation set of viewed objects, achieving state-of-the-art decoding accuracy. Copyright © 2017 Elsevier Inc. All rights reserved.
Yu, Zhen; Tan, Ee-Leng; Ni, Dong; Qin, Jing; Chen, Siping; Li, Shenli; Lei, Baiying; Wang, Tianfu
2017-05-17
Ultrasound imaging has become a prevalent examination method in prenatal diagnosis. Accurate acquisition of fetal facial standard plane (FFSP) is the most important precondition for subsequent diagnosis and measurement. In the past few years, considerable effort has been devoted to FFSP recognition using various hand-crafted features, but the recognition performance is still unsatisfactory due to the high intra-class variation of FFSPs and the high degree of visual similarity between FFSPs and other non-FFSPs. To improve the recognition performance, we propose a method to automatically recognize FFSP via a deep convolutional neural network (DCNN) architecture. The proposed DCNN consists of 16 convolutional layers with small 3×3 size kernels and three fully connected layers. A global average pooling (GAP) is adopted in the last pooling layer to significantly reduce network parameters, which alleviates the overfitting problems and improves the performance under limited training data. Both the transfer learning strategy and a data augmentation technique tailored for FFSP are implemented to further boost the recognition performance. Extensive experiments demonstrate the advantage of our proposed method over traditional approaches and the effectiveness of DCNN to recognize FFSP for clinical diagnosis.
Liu, Yan; Stojadinovic, Strahinja; Hrycushko, Brian; Wardak, Zabi; Lau, Steven; Lu, Weiguo; Yan, Yulong; Jiang, Steve B; Zhen, Xin; Timmerman, Robert; Nedzi, Lucien; Gu, Xuejun
2017-01-01
Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients' data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.
Directory of Open Access Journals (Sweden)
Yan Liu
Full Text Available Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS treatment planning. In this work, we developed a deep learning convolutional neural network (CNN algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS data and clinical patients' data. Validation on BRATS data yielded average DICE coefficients (DCs of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.
A Novel Pruning Algorithm for Smoothing Feedforward Neural Networks Based on Group Lasso Method.
Wang, Jian; Xu, Chen; Yang, Xifeng; Zurada, Jacek M
2017-09-26
In this paper, we propose four new variants of the backpropagation algorithm to improve the generalization ability for feedforward neural networks. The basic idea of these methods stems from the Group Lasso concept which deals with the variable selection problem at the group level. There are two main drawbacks when the Group Lasso penalty has been directly employed during network training. They are numerical oscillations and theoretical challenges in computing the gradients at the origin. To overcome these obstacles, smoothing functions have then been introduced by approximating the Group Lasso penalty. Numerical experiments for classification and regression problems demonstrate that the proposed algorithms perform better than the other three classical penalization methods, Weight Decay, Weight Elimination, and Approximate Smoother, on both generalization and pruning efficiency. In addition, detailed simulations based on a specific data set have been performed to compare with some other common pruning strategies, which verify the advantages of the proposed algorithm. The pruning abilities of the proposed strategy have been investigated in detail for a relatively large data set, MNIST, in terms of various smoothing approximation cases.
Artificial neural network-based pharmacogenomic algorithm for warfarin dose optimization.
Pavani, Addepalli; Naushad, Shaik Mohammad; Kumar, Rajasekar Manoj; Srinath, Murali; Malempati, Amaresh Rao; Kutala, Vijay Kumar
2016-01-01
To develop more precise pharmacogenomic algorithm for prediction of safe and effective dose of warfarin. An artificial neural network (ANN) algorithm was developed by using age, gender, BMI, plasma vitamin K levels, thyroid status and ten genetic variables as the inputs and therapeutic warfarin dose as the output. Hyperbolic tangent function was used to build an ANN architecture. This model explained 93.5% variability in warfarin dosing and predicted warfarin dose accurately in 74.5% patients whose international normalized ratio (INR) was less than 2.0 and in 83.3% patients whose INR was more than 3.5. This algorithm reduced the out-of-range INRs (odds ratio [OR]: 0.49; 95% CI: 0.30-0.79; p = 0.003), the rate of adverse drug reactions (OR: 0.00; 95% CI: 0.00-1.21; p = 0.06) and time to reach first therapeutic INR (OR: 6.73; 95% CI: 2.17-22.31; p algorithm was found to be applicable in both euthyroid and hypothyroid status. S-warfarin/7-hydroxywarfarin ratio was found to increase in subjects with CYP2C9*2 and CYP2C9*3 justifying the warfarin sensitivity attributed to these variants. An application of ANN for warfarin dosing improves predictability and provides safe and effective dosing.
EP-DNN: A Deep Neural Network-Based Global Enhancer Prediction Algorithm.
Kim, Seong Gon; Harwani, Mrudul; Grama, Ananth; Chaterji, Somali
2016-12-08
We present EP-DNN, a protocol for predicting enhancers based on chromatin features, in different cell types. Specifically, we use a deep neural network (DNN)-based architecture to extract enhancer signatures in a representative human embryonic stem cell type (H1) and a differentiated lung cell type (IMR90). We train EP-DNN using p300 binding sites, as enhancers, and TSS and random non-DHS sites, as non-enhancers. We perform same-cell and cross-cell predictions to quantify the validation rate and compare against two state-of-the-art methods, DEEP-ENCODE and RFECS. We find that EP-DNN has superior accuracy with a validation rate of 91.6%, relative to 85.3% for DEEP-ENCODE and 85.5% for RFECS, for a given number of enhancer predictions and also scales better for a larger number of enhancer predictions. Moreover, our H1 → IMR90 predictions turn out to be more accurate than IMR90 → IMR90, potentially because H1 exhibits a richer signature set and our EP-DNN model is expressive enough to extract these subtleties. Our work shows how to leverage the full expressivity of deep learning models, using multiple hidden layers, while avoiding overfitting on the training data. We also lay the foundation for exploration of cross-cell enhancer predictions, potentially reducing the need for expensive experimentation.
A Pattern Construction Scheme for Neural Network-Based Cognitive Communication
Directory of Open Access Journals (Sweden)
Ozgur Orcay
2011-01-01
Full Text Available Inefficient utilization of the frequency spectrum due to conventional regulatory limitations and physical performance limiting factors, mainly the Signal to Noise Ratio (SNR, are prominent restrictions in digital wireless communication. Pattern Based Communication System (PBCS is an adaptive and perceptual communication method based on a Cognitive Radio (CR approach. It intends an SNR oriented cognition mechanism in the physical layer for improvement of Link Spectral Efficiency (LSE. The key to this system is construction of optimal communication signals, which consist of encoded data in different pattern forms (waveforms depending on spectral availabilities. The signals distorted in the communication medium are recovered according to the pre-trained pattern glossary by the perceptual receiver. In this study, we have shown that it is possible to improve the bandwidth efficiency when largely uncorrelated signal patterns are chosen in order to form a glossary that represents symbols for different length data groups and the information can be recovered by the Artificial Neural Network (ANN in the receiver site.
Energy Technology Data Exchange (ETDEWEB)
Behler, Joerg [Department of Chemistry and Applied Biosciences, ETH Zurich, USI-Campus, Lugano (Switzerland); Lehrstuhl fuer Theoretische Chemie, Ruhr-Universitaet Bochum, 44780 Bochum (Germany); Martonak, Roman [Department of Chemistry and Applied Biosciences, ETH Zurich, USI-Campus, Lugano (Switzerland); Department of Experimental Physics, Faculty of Mathematics, Physics and Informatics, Comenius University, Mlynska dolina F2, 84248 Bratislava (Slovakia); Donadio, Davide [Department of Chemistry and Applied Biosciences, ETH Zurich, USI-Campus, Lugano (Switzerland); Department of Chemistry, UC Davis, One Shields Ave., Davis, CA 95616 (United States); Parrinello, Michele [Department of Chemistry and Applied Biosciences, ETH Zurich, USI-Campus, Lugano (Switzerland)
2008-12-15
We present a combination of the metadynamics method for the investigation of pressure-induced phase transitions in solids with a neural network representation of high-dimensional density-functional theory (DFT) potential-energy surfaces. In a recent illustration of the method for the complex high-pressure phase diagram of silicon[Behler et al., Phys. Rev. Lett. 100, 185501 (2008)] we have shown that the full sequence of phases can be reconstructed by a series of subsequent simulations. In the present paper we give a detailed account of the underlying methodology and discuss the scope and limitations of the approach, which promises to be a valuable tool for the investigation of a variety of inorganic materials. The method is several orders of magnitude faster than a direct coupling of metadynamics with electronic structure calculations, while the accuracy is essentially maintained, thus providing access to extended simulations of large systems. (copyright 2008 WILEY-VCH Verlag GmbH and Co. KGaA, Weinheim) (orig.)
Convolutional Neural Network-based SAR Image Classification with Noisy Labels
Directory of Open Access Journals (Sweden)
Zhao Juanping
2017-10-01
Full Text Available SAR image classification is an important task in SAR image interpretation. Supervised learning methods, such as the Convolutional Neural Network (CNN, demand samples that are accurately labeled. However, this presents a major challenge in SAR image labeling. Due to their unique imaging mechanism, SAR images are seriously affected by speckle, geometric distortion, and incomplete structural information. Thus, SAR images have a strong non-intuitive property, which causes difficulties in SAR image labeling, and which results in the weakened learning and generalization performance of many classifiers (including CNN. In this paper, we propose a Probability Transition CNN (PTCNN for patch-level SAR image classification with noisy labels. Based on the classical CNN, PTCNN builds a bridge between noise-free labels and their noisy versions via a noisy-label transition layer. As such, we derive a new CNN model trained with a noisily labeled training dataset that can potentially revise noisy labels and improve learning capacity with noisily labeled data. We use a 16-class land cover dataset and the MSTAR dataset to demonstrate the effectiveness of our model. Our experimental results show the PTCNN model to be robust with respect to label noise and demonstrate its promising classification performance compared with the classical CNN model. Therefore, the proposed PTCNN model could lower the standards required regarding the quality of image labels and have a variety of practical applications.
A Neural-Network-Based Nonlinear Adaptive State-Observer for Pressurized Water Reactors
Directory of Open Access Journals (Sweden)
Zhe Dong
2013-10-01
Full Text Available Although there have been some severe nuclear accidents such as Three Mile Island (USA, Chernobyl (Ukraine and Fukushima (Japan, nuclear fission energy is still a source of clean energy that can substitute for fossil fuels in a centralized way and in a great amount with commercial availability and economic competitiveness. Since the pressurized water reactor (PWR is the most widely used nuclear fission reactor, its safe, stable and efficient operation is meaningful to the current rebirth of the nuclear fission energy industry. Power-level regulation is an important technique which can deeply affect the operation stability and efficiency of PWRs. Compared with the classical power-level controllers, the advanced power-level regulators could strengthen both the closed-loop stability and control performance by feeding back the internal state-variables. However, not all of the internal state variables of a PWR can be obtained directly by measurements. To implement advanced PWR power-level control law, it is necessary to develop a state-observer to reconstruct the unmeasurable state-variables. Since a PWR is naturally a complex nonlinear system with parameters varying with power-level, fuel burnup, xenon isotope production, control rod worth and etc., it is meaningful to design a nonlinear observer for the PWR with adaptability to system uncertainties. Due to this and the strong learning capability of the multi-layer perceptron (MLP neural network, an MLP-based nonlinear adaptive observer is given for PWRs. Based upon Lyapunov stability theory, it is proved theoretically that this newly-built observer can provide bounded and convergent state-observation. This observer is then applied to the state-observation of a special PWR, i.e., the nuclear heating reactor (NHR, and numerical simulation results not only verify its feasibility but also give the relationship between the observation performance and observer parameters.
Deep neural network-based computer-assisted detection of cerebral aneurysms in MR angiography.
Nakao, Takahiro; Hanaoka, Shouhei; Nomura, Yukihiro; Sato, Issei; Nemoto, Mitsutaka; Miki, Soichiro; Maeda, Eriko; Yoshikawa, Takeharu; Hayashi, Naoto; Abe, Osamu
2017-08-24
The usefulness of computer-assisted detection (CAD) for detecting cerebral aneurysms has been reported; therefore, the improved performance of CAD will help to detect cerebral aneurysms. To develop a CAD system for intracranial aneurysms on unenhanced magnetic resonance angiography (MRA) images based on a deep convolutional neural network (CNN) and a maximum intensity projection (MIP) algorithm, and to demonstrate the usefulness of the system by training and evaluating it using a large dataset. Retrospective study. There were 450 cases with intracranial aneurysms. The diagnoses of brain aneurysms were made on the basis of MRA, which was performed as part of a brain screening program. Noncontrast-enhanced 3D time-of-flight (TOF) MRA on 3T MR scanners. In our CAD, we used a CNN classifier that predicts whether each voxel is inside or outside aneurysms by inputting MIP images generated from a volume of interest (VOI) around the voxel. The CNN was trained in advance using manually inputted labels. We evaluated our method using 450 cases with intracranial aneurysms, 300 of which were used for training, 50 for parameter tuning, and 100 for the final evaluation. Free-response receiver operating characteristic (FROC) analysis. Our CAD system detected 94.2% (98/104) of aneurysms with 2.9 false positives per case (FPs/case). At a sensitivity of 70%, the number of FPs/case was 0.26. We showed that the combination of a CNN and an MIP algorithm is useful for the detection of intracranial aneurysms. 4 Technical Efficacy Stage 1 J. Magn. Reson. Imaging 2017. © 2017 International Society for Magnetic Resonance in Medicine.
Samanta, B.; Al-Balushi, K. R.
2003-03-01
A procedure is presented for fault diagnosis of rolling element bearings through artificial neural network (ANN). The characteristic features of time-domain vibration signals of the rotating machinery with normal and defective bearings have been used as inputs to the ANN consisting of input, hidden and output layers. The features are obtained from direct processing of the signal segments using very simple preprocessing. The input layer consists of five nodes, one each for root mean square, variance, skewness, kurtosis and normalised sixth central moment of the time-domain vibration signals. The inputs are normalised in the range of 0.0 and 1.0 except for the skewness which is normalised between -1.0 and 1.0. The output layer consists of two binary nodes indicating the status of the machine—normal or defective bearings. Two hidden layers with different number of neurons have been used. The ANN is trained using backpropagation algorithm with a subset of the experimental data for known machine conditions. The ANN is tested using the remaining set of data. The effects of some preprocessing techniques like high-pass, band-pass filtration, envelope detection (demodulation) and wavelet transform of the vibration signals, prior to feature extraction, are also studied. The results show the effectiveness of the ANN in diagnosis of the machine condition. The proposed procedure requires only a few features extracted from the measured vibration data either directly or with simple preprocessing. The reduced number of inputs leads to faster training requiring far less iterations making the procedure suitable for on-line condition monitoring and diagnostics of machines.
Artificial neural network-based predictive emission monitoring system for NOx emissions
Energy Technology Data Exchange (ETDEWEB)
Ciccone, A.; Cinnamon, C.; Niejadlik, P.R. [TransCanada Energy Ltd., Toronto, ON (Canada)]|[Golder Associates, Toronto, ON (Canada)
2005-07-01
Considering the nature of long term power supply contracts that do not include mechanisms for cost recovery, developing cost-effective ways to handle changes in legislation impacting on facilities already in operation is extremely important. Also of importance is the age of the facilities, since continuous emissions monitoring (CEM) systems were not required when they were originally put into operation, but they are not yet old enough for capital stock turnover to allow for equipment changes or transition to new operations. An alternative monitoring method that is less expensive and as accurate as traditional (CEM) systems is discussed. TransCanada Energy Ltd., developed a predictive emission monitoring (PEM) system that achieved the required accuracy of the regulatory authorities using four of its gas turbine power plant facilities. Using the power plant operation variables to predict the nitric oxide (NO) portion of the exhaust emissions, the systems are founded on an artificial neural network (ANN). This paper provides a summary of the PEM system architecture and provides background information on the facilities used in the development of this approach. It was concluded that the PEM system provides a cost effective method to monitor emissions accurately and reliably at low emitting natural gas fired facilities. As well, there is a great potential for the system to be used by other industries to monitor and report emissions. The PEM system is an ideal system for the low emitting natural gas fired generating plants however the system could be adapted for other types of industries. 5 refs., 5 tabs., 2 figs.
Directory of Open Access Journals (Sweden)
Jong Hyun Kim
2017-05-01
Full Text Available Because intelligent surveillance systems have recently undergone rapid growth, research on accurately detecting humans in videos captured at a long distance is growing in importance. The existing research using visible light cameras has mainly focused on methods of human detection for daytime hours when there is outside light, but human detection during nighttime hours when there is no outside light is difficult. Thus, methods that employ additional near-infrared (NIR illuminators and NIR cameras or thermal cameras have been used. However, in the case of NIR illuminators, there are limitations in terms of the illumination angle and distance. There are also difficulties because the illuminator power must be adaptively adjusted depending on whether the object is close or far away. In the case of thermal cameras, their cost is still high, which makes it difficult to install and use them in a variety of places. Because of this, research has been conducted on nighttime human detection using visible light cameras, but this has focused on objects at a short distance in an indoor environment or the use of video-based methods to capture multiple images and process them, which causes problems related to the increase in the processing time. To resolve these problems, this paper presents a method that uses a single image captured at night on a visible light camera to detect humans in a variety of environments based on a convolutional neural network. Experimental results using a self-constructed Dongguk night-time human detection database (DNHD-DB1 and two open databases (Korea advanced institute of science and technology (KAIST and computer vision center (CVC databases, as well as high-accuracy human detection in a variety of environments, show that the method has excellent performance compared to existing methods.
Kim, Jong Hyun; Hong, Hyung Gil; Park, Kang Ryoung
2017-05-08
Because intelligent surveillance systems have recently undergone rapid growth, research on accurately detecting humans in videos captured at a long distance is growing in importance. The existing research using visible light cameras has mainly focused on methods of human detection for daytime hours when there is outside light, but human detection during nighttime hours when there is no outside light is difficult. Thus, methods that employ additional near-infrared (NIR) illuminators and NIR cameras or thermal cameras have been used. However, in the case of NIR illuminators, there are limitations in terms of the illumination angle and distance. There are also difficulties because the illuminator power must be adaptively adjusted depending on whether the object is close or far away. In the case of thermal cameras, their cost is still high, which makes it difficult to install and use them in a variety of places. Because of this, research has been conducted on nighttime human detection using visible light cameras, but this has focused on objects at a short distance in an indoor environment or the use of video-based methods to capture multiple images and process them, which causes problems related to the increase in the processing time. To resolve these problems, this paper presents a method that uses a single image captured at night on a visible light camera to detect humans in a variety of environments based on a convolutional neural network. Experimental results using a self-constructed Dongguk night-time human detection database (DNHD-DB1) and two open databases (Korea advanced institute of science and technology (KAIST) and computer vision center (CVC) databases), as well as high-accuracy human detection in a variety of environments, show that the method has excellent performance compared to existing methods.
Neural network based control of an absorption column in the process of bioethanol production
Directory of Open Access Journals (Sweden)
Eduardo Eyng
2009-08-01
Full Text Available Gaseous ethanol may be recovered from the effluent gas mixture of the sugar cane fermentation process using a staged absorption column. In the present work, the development of a nonlinear controller, based on a neural network inverse model (ANN controller, was proposed and tested to manipulate the absorbent flow rate in order to control the residual ethanol concentration in the effluent gas phase. Simulation studies were carried out, in which a noise was applied to the ethanol concentration signals from the rigorous model. The ANN controller outperformed the dynamic matrix control (DMC when step disturbances were imposed to the gas mixture composition. A security device, based on a conventional feedback algorithm, and a digital filter were added to the proposed strategy to improve the system robustness when unforeseen operating and environmental conditions occured. The results demonstrated that ANN controller was a robust and reliable tool to control the absorption column.Deseja-se recuperar o etanol perdido por evaporação durante o processo de fermentação da cana-de-açúcar. Para tanto, faz-se uso de uma coluna de absorção. O controle da concentração de etanol no efluente gasoso da coluna é realizado pela manipulação da vazão de solvente, sendo esta determinada pelo controlador não linear proposto, baseado em um modelo inverso de redes neurais (controlador ANN. Foram feitas simulações adicionando-se um sinal de ruído a medida de concentração de etanol na fase gasosa. Quando perturbações degrau foram inseridas na mistura gasosa afluente, o controlador ANN demonstrou desempenho superior ao controle por matriz dinâmica (DMC. Um dispositivo de segurança, baseado em um controlador feedback convencional, e um filtro digital foram implementados à estratégia de controle proposta para agregar robustez no tratamento de distúrbios ocorridos no ambiente operacional. Os resultados demonstraram que o controlador ANN é uma
Fernández-Alemán, José Luis; López-González, Laura; González-Sequeros, Ofelia; Jayne, Chrisina; López-Jiménez, Juan José; Carrillo-de-Gea, Juan Manuel; Toval, Ambrosio
2016-04-01
This paper presents an empirical study of a formative neural network-based assessment approach by using mobile technology to provide pharmacy students with intelligent diagnostic feedback. An unsupervised learning algorithm was integrated with an audience response system called SIDRA in order to generate states that collect some commonality in responses to questions and add diagnostic feedback for guided learning. A total of 89 pharmacy students enrolled on a Human Anatomy course were taught using two different teaching methods. Forty-four students employed intelligent SIDRA (i-SIDRA), whereas 45 students received the same training but without using i-SIDRA. A statistically significant difference was found between the experimental group (i-SIDRA) and the control group (traditional learning methodology), with T (87) = 6.598, p SIDRA and the methodology used during the process of learning anatomy (M = 4.59). The new empirical contribution presented in this paper allows instructors to perform post hoc analyses of each particular student's progress to ensure appropriate training.
Directory of Open Access Journals (Sweden)
Lund Ole
2009-09-01
Full Text Available Abstract Background The major histocompatibility complex (MHC molecule plays a central role in controlling the adaptive immune response to infections. MHC class I molecules present peptides derived from intracellular proteins to cytotoxic T cells, whereas MHC class II molecules stimulate cellular and humoral immunity through presentation of extracellularly derived peptides to helper T cells. Identification of which peptides will bind a given MHC molecule is thus of great importance for the understanding of host-pathogen interactions, and large efforts have been placed in developing algorithms capable of predicting this binding event. Results Here, we present a novel artificial neural network-based method, NN-align that allows for simultaneous identification of the MHC class II binding core and binding affinity. NN-align is trained using a novel training algorithm that allows for correction of bias in the training data due to redundant binding core representation. Incorporation of information about the residues flanking the peptide-binding core is shown to significantly improve the prediction accuracy. The method is evaluated on a large-scale benchmark consisting of six independent data sets covering 14 human MHC class II alleles, and is demonstrated to outperform other state-of-the-art MHC class II prediction methods. Conclusion The NN-align method is competitive with the state-of-the-art MHC class II peptide binding prediction algorithms. The method is publicly available at http://www.cbs.dtu.dk/services/NetMHCII-2.0.
Ansari, Nirwan; Liu, Dequan
1991-01-01
A neural-network-based traffic management scheme for a satellite communication network is described. The scheme consists of two levels of management. The front end of the scheme is a derivation of Kohonen's self-organization model to configure maps for the satellite communication network dynamically. The model consists of three stages. The first stage is the pattern recognition task, in which an exemplar map that best meets the current network requirements is selected. The second stage is the analysis of the discrepancy between the chosen exemplar map and the state of the network, and the adaptive modification of the chosen exemplar map to conform closely to the network requirement (input data pattern) by means of Kohonen's self-organization. On the basis of certain performance criteria, whether a new map is generated to replace the original chosen map is decided in the third stage. A state-dependent routing algorithm, which arranges the incoming call to some proper path, is used to make the network more efficient and to lower the call block rate. Simulation results demonstrate that the scheme, which combines self-organization and the state-dependent routing mechanism, provides better performance in terms of call block rate than schemes that only have either the self-organization mechanism or the routing mechanism.
Monfared, Vahid
2017-03-01
The present work presents a new approach based on neural network prediction for simple and fast estimation of the creep plastic behaviour of the short fiber composites. Also, this approach is proposed to reduce the solution procedure. Moreover, as a significant application of the method, shuttles and spaceships, turbine blades and discs are generally subjected to the creep effects. Consequently, analysis of the creep phenomenon is required and vital in different industries. Analysis of the creep behaviour is required for failure, fracture, fatigue, and creep resistance of the optoelectronic/photonic composites, and sensors. One of the main applications of the present work is in designing the composites with optical fibers and devices. At last, a good agreement is seen among the present prediction by neural network approach, finite element method (FEM), and the experimental results.
Sauzède, R.; Claustre, H.; Uitz, J.; Jamet, C.; Dall'Olmo, G.; D'Ortenzio, F.; Gentili, B.; Poteau, A.; Schmechtig, C.
2016-04-01
The present study proposes a novel method that merges satellite ocean color bio-optical products with Argo temperature-salinity profiles to infer the vertical distribution of the particulate backscattering coefficient (bbp). This neural network-based method (SOCA-BBP for Satellite Ocean-Color merged with Argo data to infer the vertical distribution of the Particulate Backscattering coefficient) uses three main input components: (1) satellite-based surface estimates of bbp and chlorophyll a concentration matched up in space and time with (2) depth-resolved physical properties derived from temperature-salinity profiles measured by Argo profiling floats and (3) the day of the year of the considered satellite-Argo matchup. The neural network is trained and validated using a database including 4725 simultaneous profiles of temperature-salinity and bio-optical properties collected by Bio-Argo floats, with concomitant satellite-derived products. The Bio-Argo profiles are representative of the global open-ocean in terms of oceanographic conditions, making the proposed method applicable to most open-ocean environments. SOCA-BBP is validated using 20% of the entire database (global error of 21%). We present additional validation results based on two other independent data sets acquired (1) by four Bio-Argo floats deployed in major oceanic basins, not represented in the database used to train the method; and (2) during an AMT (Atlantic Meridional Transect) field cruise in 2009. These validation tests based on two fully independent data sets indicate the robustness of the predicted vertical distribution of bbp. To illustrate the potential of the method, we merged monthly climatological Argo profiles with ocean color products to produce a depth-resolved climatology of bbp for the global ocean.
Tsehay, Yohannes K.; Lay, Nathan S.; Roth, Holger R.; Wang, Xiaosong; Kwak, Jin Tae; Turkbey, Baris I.; Pinto, Peter A.; Wood, Brad J.; Summers, Ronald M.
2017-03-01
Prostate cancer (PCa) is the second most common cause of cancer related deaths in men. Multiparametric MRI (mpMRI) is the most accurate imaging method for PCa detection; however, it requires the expertise of experienced radiologists leading to inconsistency across readers of varying experience. To increase inter-reader agreement and sensitivity, we developed a computer-aided detection (CAD) system that can automatically detect lesions on mpMRI that readers can use as a reference. We investigated a convolutional neural network based deep-learing (DCNN) architecture to find an improved solution for PCa detection on mpMRI. We adopted a network architecture from a state-of-the-art edge detector that takes an image as an input and produces an image probability map. Two-fold cross validation along with a receiver operating characteristic (ROC) analysis and free-response ROC (FROC) were used to determine our deep-learning based prostate-CAD's (CADDL) performance. The efficacy was compared to an existing prostate CAD system that is based on hand-crafted features, which was evaluated on the same test-set. CADDL had an 86% detection rate at 20% false-positive rate while the top-down learning CAD had 80% detection rate at the same false-positive rate, which translated to 94% and 85% detection rate at 10 false-positives per patient on the FROC. A CNN based CAD is able to detect cancerous lesions on mpMRI of the prostate with results comparable to an existing prostate-CAD showing potential for further development.
Directory of Open Access Journals (Sweden)
Olha V. Zastelo
2015-09-01
Full Text Available In this article the method of the integral assessment of the level of students` foreign language communicative competence by the group of experts through the complex test in a foreign language is considered. The use of mathematical methods and modern specialized software during complex testing of students significantly improves the expert methods, particularly in the direction of increasing the reliability of the assessment. Capitalizing analytical software environment realizes the simulation of non-linear generalizations based on artificial neural networks, which increases the accuracy of the estimate and allows further efficient use of the competent experts` experience gained in the model.
Neural network-based robust actuator fault diagnosis for a non-linear multi-tank system.
Mrugalski, Marcin; Luzar, Marcel; Pazera, Marcin; Witczak, Marcin; Aubrun, Christophe
2016-03-01
The paper is devoted to the problem of the robust actuator fault diagnosis of the dynamic non-linear systems. In the proposed method, it is assumed that the diagnosed system can be modelled by the recurrent neural network, which can be transformed into the linear parameter varying form. Such a system description allows developing the designing scheme of the robust unknown input observer within H∞ framework for a class of non-linear systems. The proposed approach is designed in such a way that a prescribed disturbance attenuation level is achieved with respect to the actuator fault estimation error, while guaranteeing the convergence of the observer. The application of the robust unknown input observer enables actuator fault estimation, which allows applying the developed approach to the fault tolerant control tasks. Copyright © 2016 ISA. Published by Elsevier Ltd. All rights reserved.
Application of radial basis neural network for state estimation of ...
African Journals Online (AJOL)
user
An original application of radial basis function (RBF) neural network for power system state estimation is proposed in this paper. The property of massive parallelism of neural networks is employed for this. The application of RBF neural network for state estimation is investigated by testing its applicability on a IEEE 14 bus ...
A dynamic programming approach for quickly estimating large network-based MEV models
DEFF Research Database (Denmark)
Mai, Tien; Frejinger, Emma; Fosgerau, Mogens
2017-01-01
We propose a way to estimate a family of static Multivariate Extreme Value (MEV) models with large choice sets in short computational time. The resulting model is also straightforward and fast to use for prediction. Following Daly and Bierlaire (2006), the correlation structure is defined by a ro...... to converge (4.3 h on an Intel(R) 3.2 GHz machine using a non-parallelized code). We also show that our approach allows to estimate a cross-nested logit model of 111 nests with a real data set of more than 100,000 observations in 14 h.......We propose a way to estimate a family of static Multivariate Extreme Value (MEV) models with large choice sets in short computational time. The resulting model is also straightforward and fast to use for prediction. Following Daly and Bierlaire (2006), the correlation structure is defined...... by a rooted, directed graph where each node without successor is an alternative. We formulate a family of MEV models as dynamic discrete choice models on graphs of correlation structures and show that the dynamic models are consistent with MEV theory and generalize the network MEV model (Daly and Bierlaire...
A network-based approach for estimating pedestrian journey-time exposure to air pollution.
Davies, Gemma; Whyatt, J Duncan
2014-07-01
Individual exposure to air pollution depends not only upon pollution concentrations in the surrounding environment, but also on the volume of air inhaled, which is determined by an individual's physiology and activity level. This study focuses on journey-time exposure, using network analysis in a GIS environment to identify pedestrian routes between multiple origins and destinations throughout the city of Lancaster, North West England. For each segment of a detailed footpath network, exposure was calculated accounting for PM2.5 concentrations (estimated using an atmospheric dispersion model) and respiratory minute volume (varying between individuals and with slope). For each of the routes generated the cumulative exposure to PM2.5 was estimated, allowing for easy comparison between multiple routes. Significant variations in exposure were found between routes depending on their geography, as well as in response to variations in background concentrations and meteorology between days. Differences in physiological characteristics such as age or weight were also seen to impact journey-time exposure considerably. In addition to assessing exposure for a given route, the approach was used to identify alternative routes that minimised journey-time exposure. Exposure reduction potential varied considerably between days, with even subtle shifts in route location, such as to the opposite side of the road, showing significant benefits. The method presented is both flexible and scalable, allowing for the interactions between physiology, activity level, pollution concentration and journey duration to be explored. In enabling physiology and activity level to be integrated into exposure calculations a more comprehensive estimate of journey-time exposure can be made, which has potential to provide more realistic inputs for epidemiological studies. Copyright © 2014 Elsevier B.V. All rights reserved.
National Research Council Canada - National Science Library
Kerdphol, Thongchart; Qudaih, Yaser; Watanabe, Masayuki; Mitani, Yasunori
2016-01-01
...) in a short period of time.This paper presents a new method for the intelligent online management of both active and reactive power of a BESS based on a radial basis function neural network (RBFNN...
Modeling and Density Estimation of an Urban Freeway Network Based on Dynamic Graph Hybrid Automata.
Chen, Yangzhou; Guo, Yuqi; Wang, Ying
2017-03-29
In this paper, in order to describe complex network systems, we firstly propose a general modeling framework by combining a dynamic graph with hybrid automata and thus name it Dynamic Graph Hybrid Automata (DGHA). Then we apply this framework to model traffic flow over an urban freeway network by embedding the Cell Transmission Model (CTM) into the DGHA. With a modeling procedure, we adopt a dual digraph of road network structure to describe the road topology, use linear hybrid automata to describe multi-modes of dynamic densities in road segments and transform the nonlinear expressions of the transmitted traffic flow between two road segments into piecewise linear functions in terms of multi-mode switchings. This modeling procedure is modularized and rule-based, and thus is easily-extensible with the help of a combination algorithm for the dynamics of traffic flow. It can describe the dynamics of traffic flow over an urban freeway network with arbitrary topology structures and sizes. Next we analyze mode types and number in the model of the whole freeway network, and deduce a Piecewise Affine Linear System (PWALS) model. Furthermore, based on the PWALS model, a multi-mode switched state observer is designed to estimate the traffic densities of the freeway network, where a set of observer gain matrices are computed by using the Lyapunov function approach. As an example, we utilize the PWALS model and the corresponding switched state observer to traffic flow over Beijing third ring road. In order to clearly interpret the principle of the proposed method and avoid computational complexity, we adopt a simplified version of Beijing third ring road. Practical application for a large-scale road network will be implemented by decentralized modeling approach and distributed observer designing in the future research.
Guarneri, Paolo; Rocca, Gianpiero; Gobbi, Massimiliano
2008-09-01
This paper deals with the simulation of the tire/suspension dynamics by using recurrent neural networks (RNNs). RNNs are derived from the multilayer feedforward neural networks, by adding feedback connections between output and input layers. The optimal network architecture derives from a parametric analysis based on the optimal tradeoff between network accuracy and size. The neural network can be trained with experimental data obtained in the laboratory from simulated road profiles (cleats). The results obtained from the neural network demonstrate good agreement with the experimental results over a wide range of operation conditions. The NN model can be effectively applied as a part of vehicle system model to accurately predict elastic bushings and tire dynamics behavior. Although the neural network model, as a black-box model, does not provide a good insight of the physical behavior of the tire/suspension system, it is a useful tool for assessing vehicle ride and noise, vibration, harshness (NVH) performance due to its good computational efficiency and accuracy.
Advantages of Neural Network Based Air Data Estimation for Unmanned Aerial Vehicles
Angelo Lerro; Manuela Battipede; Piero Gili; Alberto Brandl
2017-01-01
Redundancy requirements for UAV (Unmanned Aerial Vehicle) are hardly faced due to the generally restricted amount of available space and allowable weight for the aircraft systems, limiting their exploitation. Essential equipment as the Air Data, Attitude and Heading Reference Systems (ADAHRS) require several external probes to measure significant data as the Angle of Attack or the Sideslip Angle. Previous research focused on the analysis of a patented technology named Smart-ADAHRS (Smart Air ...
Directory of Open Access Journals (Sweden)
T. D. Xenos
2002-01-01
Full Text Available In this work, Neural-Network-based single-station hourly daily foF2 and M(3000F2 modelling of 15 European ionospheric stations is investigated. The data used are neural networks and hourly daily values from the period 1964- 1988 for training the neural networks and from the period 1989-1994 for checking the prediction accuracy. Two types of models are presented for the F2-layer critical frequency prediction and two for the propagation factor M(3000F2. The first foF2 model employs the E-layer local noon calculated daily critical frequency (foE12 and the local noon F2- layer critical frequency of the previous day. The second foF2 model, which introduces a new regional mapping technique, employs the Juliusruh neural network model and uses the E-layer local noon calculated daily critical frequency (foE12, and the previous day F2-layer critical frequency measured at Juliusruh at noon. The first M(3000F2 model employs the E-layer local noon calculated daily critical frequency (foE12, its ± 3 h deviations and the local noon cosine of the solar zenith angle (cos c12. The second model, which introduces a new M(3000F2 mapping technique, employs Juliusruh neural network model and uses the E-layer local noon calculated daily critical frequency (foE12, and the previous day F2-layer critical frequency measured at Juliusruh at noon.
Mapping Neural Network Derived from the Parzen Window Estimator
DEFF Research Database (Denmark)
Schiøler, Henrik; Hartmann, U.
1992-01-01
The article presents a general theoretical basis for the construction of mapping neural networks. The theory is based on the Parzen Window estimator for......The article presents a general theoretical basis for the construction of mapping neural networks. The theory is based on the Parzen Window estimator for...
Liu, Zhijian; Liu, Kejun; Li, Hao; Zhang, Xinyu; Jin, Guangya; Cheng, Kewei
2015-01-01
Measurements of heat collection rate and heat loss coefficient are crucial for the evaluation of in service water-in-glass evacuated tube solar water heaters. However, conventional measurement requires expensive detection devices and undergoes a series of complicated procedures. To simplify the measurement and reduce the cost, software based on artificial neural networks for measuring heat collection rate and heat loss coefficient of water-in-glass evacuated tube solar water heaters was developed. Using multilayer feed-forward neural networks with back-propagation algorithm, we developed and tested our program on the basis of 915 measured samples of water-in-glass evacuated tube solar water heaters. This artificial neural networks-based software program automatically obtained accurate heat collection rate and heat loss coefficient using simply "portable test instruments" acquired parameters, including tube length, number of tubes, tube center distance, heat water mass in tank, collector area, angle between tubes and ground and final temperature. Our results show that this software (on both personal computer and Android platforms) is efficient and convenient to predict the heat collection rate and heat loss coefficient due to it slow root mean square errors in prediction. The software now can be downloaded from http://t.cn/RLPKF08.
Talebi, H A; Khorasani, K; Tafazoli, S
2009-01-01
This paper presents a robust fault detection and isolation (FDI) scheme for a general class of nonlinear systems using a neural-network-based observer strategy. Both actuator and sensor faults are considered. The nonlinear system considered is subject to both state and sensor uncertainties and disturbances. Two recurrent neural networks are employed to identify general unknown actuator and sensor faults, respectively. The neural network weights are updated according to a modified backpropagation scheme. Unlike many previous methods developed in the literature, our proposed FDI scheme does not rely on availability of full state measurements. The stability of the overall FDI scheme in presence of unknown sensor and actuator faults as well as plant and sensor noise and uncertainties is shown by using the Lyapunov's direct method. The stability analysis developed requires no restrictive assumptions on the system and/or the FDI algorithm. Magnetorquer-type actuators and magnetometer-type sensors that are commonly employed in the attitude control subsystem (ACS) of low-Earth orbit (LEO) satellites for attitude determination and control are considered in our case studies. The effectiveness and capabilities of our proposed fault diagnosis strategy are demonstrated and validated through extensive simulation studies.
Directory of Open Access Journals (Sweden)
Fei Chen
2013-01-01
Full Text Available This paper deals with the finite-time stabilization problem for discrete-time Markov jump nonlinear systems with time delays and norm-bounded exogenous disturbance. The nonlinearities in different jump modes are parameterized by neural networks. Subsequently, a linear difference inclusion state space representation for a class of neural networks is established. Based on this, sufficient conditions are derived in terms of linear matrix inequalities to guarantee stochastic finite-time boundedness and stochastic finite-time stabilization of the closed-loop system. A numerical example is illustrated to verify the efficiency of the proposed technique.
Li, Xuanying; Li, Xiaotong; Hu, Cheng
2017-12-01
In this paper, without transforming the second order inertial neural networks into the first order differential systems by some variable substitutions, asymptotic stability and synchronization for a class of delayed inertial neural networks are investigated. Firstly, a new Lyapunov functional is constructed to directly propose the asymptotic stability of the inertial neural networks, and some new stability criteria are derived by means of Barbalat Lemma. Additionally, by designing a new feedback control strategy, the asymptotic synchronization of the addressed inertial networks is studied and some effective conditions are obtained. To reduce the control cost, an adaptive control scheme is designed to realize the asymptotic synchronization. It is noted that the dynamical behaviors of inertial neural networks are directly analyzed in this paper by constructing some new Lyapunov functionals, this is totally different from the traditional reduced-order variable substitution method. Finally, some numerical simulations are given to demonstrate the effectiveness of the derived theoretical results. Copyright © 2017 Elsevier Ltd. All rights reserved.
DEFF Research Database (Denmark)
Săftoiu, Adrian; Vilmann, Peter; Gorunescu, Florin
2012-01-01
By using strain assessment, real-time endoscopic ultrasound (EUS) elastography provides additional information about a lesion's characteristics in the pancreas. We assessed the accuracy of real-time EUS elastography in focal pancreatic lesions using computer-aided diagnosis by artificial neural...
Acernese, F; Rosa, M D; Rosa, R D; Eleuteri, A; Milano, L; Tagliaferri, R
2002-01-01
In this paper, a neural network-based approach is presented for the real time noise identification of a GW laser interferometric antenna. The 40 m Caltech laser interferometer output data provide a realistic test bed for noise identification algorithms because of the presence of many relevant effects: violin resonances in the suspensions, main power harmonics, ring-down noise from servo control systems, electronic noises, glitches and so on. These effects can be assumed to be present in all the first interferometric long baseline GW antennas such as VIRGO, LIGO, GEO and TAMA. For noise identification, we used the Caltech-40 m laser interferometer data. The results we obtained are pretty good notwithstanding the high initial computational cost. The algorithm we propose is general and robust, taking into account that it does not require a priori information on the data, nor a precise model, and it constitutes a powerful tool for time series data analysis.
Energy Technology Data Exchange (ETDEWEB)
Acernese, F [Dipartimento di Scienze Fisiche, Universita di Napoli Federico II, Complesso Universitario di Monte S Angelo, via Cintia, I-80126 Naples (Italy); Barone, F [Istituto Nazionale di Fisica Nucleare, sez. Napoli, Complesso Universitario di Monte S Angelo, via Cintia, I-80126 Naples (Italy); Rosa, M de [Dipartimento di Scienze Fisiche, Universita di Napoli Federico II, Complesso Universitario di Monte S Angelo, via Cintia, I-80126 Naples (Italy); Rosa, R De [Dipartimento di Scienze Fisiche, Universita di Napoli Federico II, Complesso Universitario di Monte S Angelo, via Cintia, I-80126 Naples (Italy); Eleuteri, A [Istituto Nazionale di Fisica Nucleare, sez. Napoli, Complesso Universitario di Monte S Angelo, via Cintia, I-80126 Naples (Italy); Milano, L [Dipartimento di Scienze Fisiche, Universita di Napoli Federico II, Complesso Universitario di Monte S Angelo, via Cintia, I-80126 Naples (Italy); Tagliaferri, R [Dipartimento di Matematica ed Informatica, Universita di Salerno, via S Allende, I-84081 Baronissi (Salerno) (Italy)
2002-06-21
In this paper, a neural network-based approach is presented for the real time noise identification of a GW laser interferometric antenna. The 40 m Caltech laser interferometer output data provide a realistic test bed for noise identification algorithms because of the presence of many relevant effects: violin resonances in the suspensions, main power harmonics, ring-down noise from servo control systems, electronic noises, glitches and so on. These effects can be assumed to be present in all the first interferometric long baseline GW antennas such as VIRGO, LIGO, GEO and TAMA. For noise identification, we used the Caltech-40 m laser interferometer data. The results we obtained are pretty good notwithstanding the high initial computational cost. The algorithm we propose is general and robust, taking into account that it does not require a priori information on the data, nor a precise model, and it constitutes a powerful tool for time series data analysis.
A neural network-based method for spectral distortion correction in photon counting x-ray CT
Touch, Mengheng; Clark, Darin P.; Barber, William; Badea, Cristian T.
2016-08-01
Spectral CT using a photon counting x-ray detector (PCXD) shows great potential for measuring material composition based on energy dependent x-ray attenuation. Spectral CT is especially suited for imaging with K-edge contrast agents to address the otherwise limited contrast in soft tissues. We have developed a micro-CT system based on a PCXD. This system enables both 4 energy bins acquisition, as well as full-spectrum mode in which the energy thresholds of the PCXD are swept to sample the full energy spectrum for each detector element and projection angle. Measurements provided by the PCXD, however, are distorted due to undesirable physical effects in the detector and can be very noisy due to photon starvation in narrow energy bins. To address spectral distortions, we propose and demonstrate a novel artificial neural network (ANN)-based spectral distortion correction mechanism, which learns to undo the distortion in spectral CT, resulting in improved material decomposition accuracy. To address noise, post-reconstruction denoising based on bilateral filtration, which jointly enforces intensity gradient sparsity between spectral samples, is used to further improve the robustness of ANN training and material decomposition accuracy. Our ANN-based distortion correction method is calibrated using 3D-printed phantoms and a model of our spectral CT system. To enable realistic simulations and validation of our method, we first modeled the spectral distortions using experimental data acquired from 109Cd and 133Ba radioactive sources measured with our PCXD. Next, we trained an ANN to learn the relationship between the distorted spectral CT projections and the ideal, distortion-free projections in a calibration step. This required knowledge of the ground truth, distortion-free spectral CT projections, which were obtained by simulating a spectral CT scan of the digital version of a 3D-printed phantom. Once the training was completed, the trained ANN was used to perform
Jung, Jun-Young; Heo, Wonho; Yang, Hyundae; Park, Hyunsub
2015-01-01
An exact classification of different gait phases is essential to enable the control of exoskeleton robots and detect the intentions of users. We propose a gait phase classification method based on neural networks using sensor signals from lower limb exoskeleton robots. In such robots, foot sensors with force sensing registers are commonly used to classify gait phases. We describe classifiers that use the orientation of each lower limb segment and the angular velocities of the joints to output the current gait phase. Experiments to obtain the input signals and desired outputs for the learning and validation process are conducted, and two neural network methods (a multilayer perceptron and nonlinear autoregressive with external inputs (NARX)) are used to develop an optimal classifier. Offline and online evaluations using four criteria are used to compare the performance of the classifiers. The proposed NARX-based method exhibits sufficiently good performance to replace foot sensors as a means of classifying gait phases. PMID:26528986
Directory of Open Access Journals (Sweden)
Jun-Young Jung
2015-10-01
Full Text Available An exact classification of different gait phases is essential to enable the control of exoskeleton robots and detect the intentions of users. We propose a gait phase classification method based on neural networks using sensor signals from lower limb exoskeleton robots. In such robots, foot sensors with force sensing registers are commonly used to classify gait phases. We describe classifiers that use the orientation of each lower limb segment and the angular velocities of the joints to output the current gait phase. Experiments to obtain the input signals and desired outputs for the learning and validation process are conducted, and two neural network methods (a multilayer perceptron and nonlinear autoregressive with external inputs (NARX are used to develop an optimal classifier. Offline and online evaluations using four criteria are used to compare the performance of the classifiers. The proposed NARX-based method exhibits sufficiently good performance to replace foot sensors as a means of classifying gait phases.
Jung, Jun-Young; Heo, Wonho; Yang, Hyundae; Park, Hyunsub
2015-10-30
An exact classification of different gait phases is essential to enable the control of exoskeleton robots and detect the intentions of users. We propose a gait phase classification method based on neural networks using sensor signals from lower limb exoskeleton robots. In such robots, foot sensors with force sensing registers are commonly used to classify gait phases. We describe classifiers that use the orientation of each lower limb segment and the angular velocities of the joints to output the current gait phase. Experiments to obtain the input signals and desired outputs for the learning and validation process are conducted, and two neural network methods (a multilayer perceptron and nonlinear autoregressive with external inputs (NARX)) are used to develop an optimal classifier. Offline and online evaluations using four criteria are used to compare the performance of the classifiers. The proposed NARX-based method exhibits sufficiently good performance to replace foot sensors as a means of classifying gait phases.
Ventura, Cristina; Latino, Diogo A R S; Martins, Filomena
2013-01-01
The performance of two QSAR methodologies, namely Multiple Linear Regressions (MLR) and Neural Networks (NN), towards the modeling and prediction of antitubercular activity was evaluated and compared. A data set of 173 potentially active compounds belonging to the hydrazide family and represented by 96 descriptors was analyzed. Models were built with Multiple Linear Regressions (MLR), single Feed-Forward Neural Networks (FFNNs), ensembles of FFNNs and Associative Neural Networks (AsNNs) using four different data sets and different types of descriptors. The predictive ability of the different techniques used were assessed and discussed on the basis of different validation criteria and results show in general a better performance of AsNNs in terms of learning ability and prediction of antitubercular behaviors when compared with all other methods. MLR have, however, the advantage of pinpointing the most relevant molecular characteristics responsible for the behavior of these compounds against Mycobacterium tuberculosis. The best results for the larger data set (94 compounds in training set and 18 in test set) were obtained with AsNNs using seven descriptors (R(2) of 0.874 and RMSE of 0.437 against R(2) of 0.845 and RMSE of 0.472 in MLRs, for test set). Counter-Propagation Neural Networks (CPNNs) were trained with the same data sets and descriptors. From the scrutiny of the weight levels in each CPNN and the information retrieved from MLRs, a rational design of potentially active compounds was attempted. Two new compounds were synthesized and tested against M. tuberculosis showing an activity close to that predicted by the majority of the models. Copyright © 2013 Elsevier Masson SAS. All rights reserved.
Directory of Open Access Journals (Sweden)
Radi Radi
2011-08-01
Full Text Available Constructive Back Propagation Neural Network (CBPNN is a kind of back propagation neural network trained with constructive algorithm. Training of CBPNN is mainly conducted by developing the network’s architecture which commonly done by adding a number of new neuron units on learning process. Training of the network usually implements fixed method to develop its structure gradually by adding new units constantly. Although this method is simple and able to create an adaptive network for data pattern complexity, but it is wasteful and inefficient for computing. New unit addition affects directly to the computational load of training, speed of convergence, and structure of the final neural network. While increases training load significantly, excessive addition of units also tends to generate a large size of final network. Moreover, addition pattern with small unit number tends to drop off the adaptability of the network and extends time of training. Therefore, there is important to design an adaptive structure development pattern for CBPNN in order to minimize computing load of training. This study proposes Fuzzy Logic (FL algorithm to manage and develop structure of CBPNN. FL method was implemented on two models of CBPNN, i.e. designed with one and two hidden layers, used to recognize aroma patterns on an electronic nose system. The results showed that this method is effective to be applied due to its capability to minimize time of training, to reduce load of computational learning, and generate small size of network.
Mandal, Sudip; Khan, Abhinandan; Saha, Goutam; Pal, Rajat K
2016-01-01
The accurate prediction of genetic networks using computational tools is one of the greatest challenges in the postgenomic era. Recurrent Neural Network is one of the most popular but simple approaches to model the network dynamics from time-series microarray data. To date, it has been successfully applied to computationally derive small-scale artificial and real-world genetic networks with high accuracy. However, they underperformed for large-scale genetic networks. Here, a new methodology has been proposed where a hybrid Cuckoo Search-Flower Pollination Algorithm has been implemented with Recurrent Neural Network. Cuckoo Search is used to search the best combination of regulators. Moreover, Flower Pollination Algorithm is applied to optimize the model parameters of the Recurrent Neural Network formalism. Initially, the proposed method is tested on a benchmark large-scale artificial network for both noiseless and noisy data. The results obtained show that the proposed methodology is capable of increasing the inference of correct regulations and decreasing false regulations to a high degree. Secondly, the proposed methodology has been validated against the real-world dataset of the DNA SOS repair network of Escherichia coli. However, the proposed method sacrifices computational time complexity in both cases due to the hybrid optimization process.
Directory of Open Access Journals (Sweden)
Sudip Mandal
2016-01-01
Full Text Available The accurate prediction of genetic networks using computational tools is one of the greatest challenges in the postgenomic era. Recurrent Neural Network is one of the most popular but simple approaches to model the network dynamics from time-series microarray data. To date, it has been successfully applied to computationally derive small-scale artificial and real-world genetic networks with high accuracy. However, they underperformed for large-scale genetic networks. Here, a new methodology has been proposed where a hybrid Cuckoo Search-Flower Pollination Algorithm has been implemented with Recurrent Neural Network. Cuckoo Search is used to search the best combination of regulators. Moreover, Flower Pollination Algorithm is applied to optimize the model parameters of the Recurrent Neural Network formalism. Initially, the proposed method is tested on a benchmark large-scale artificial network for both noiseless and noisy data. The results obtained show that the proposed methodology is capable of increasing the inference of correct regulations and decreasing false regulations to a high degree. Secondly, the proposed methodology has been validated against the real-world dataset of the DNA SOS repair network of Escherichia coli. However, the proposed method sacrifices computational time complexity in both cases due to the hybrid optimization process.
Directory of Open Access Journals (Sweden)
Dan ZHANG
2010-10-01
Full Text Available Accuracy is paramount for the further development of parallel mechanism in real world, especially in industry. Previous research was focused on the improvement of rigidity and load capacity which is related with the stiffness matrix of closed loop kinematic structure. However, if the mechanical structure has been predefined or fabricated, stiffness matrix only can search for the optimal configuration in the workspace, but fails to enhance the accuracy at a given pose. In this research, the concept of optimal robust calibration is developed as an effective approach to largely reduce various errors of the predefined parallel mechanism. A novel coevolutionary radial basis function (RBF neural network based soft sensor is proposed to implement the optimal robust calibration procedure. A six- degrees-of-freedom parallel kinematics manipulator is selected as the case study to verify the proposed methodology. The results demonstrate that the coevolutionary neural network possesses the excellent performance as a smart soft sensor for the calibration of closed loop kinematic structure.
ESTIMATION OF PV MODULE SURFACE TEMPERATURE USING ARTIFICIAL NEURAL NETWORKS
Directory of Open Access Journals (Sweden)
Can Coskun
2016-12-01
Full Text Available This study aimed to use the artificial neural network (ANN method to estimate the surface temperature of a photovoltaic (PV panel. Using the experimentally obtained PV data, the accuracy of the ANN model was evaluated. To train the artificial neural network (ANN, outer temperature solar radiation and wind speed values were inputs and surface temperature was an output. The ANN was used to estimate PV panel surface temperature. Using the Levenberg-Marquardt (LM algorithm the feed forward artificial neural network was trained. Two back propagation type ANN algorithms were used and their performance was compared with the estimate from the LM algorithm. To train the artificial neural network, experimental data were used for two thirds with the remaining third used for testing. Additionally scaled conjugate gradient (SCG back propagation and resilient back propagation (RB type ANN algorithms were used for comparison with the LM algorithm. The performances of these three types of artificial neural network were compared and mean error rates of between 0.005962 and 0.012177% were obtained. The best estimate was produced by the LM algorithm. Estimation of PV surface temperature with artificial neural networks provides better results than conventional correlation methods. This study showed that artificial neural networks may be effectively used to estimate PV surface temperature.
Bose, Subrata K.; Browne, Antony; Kazemian, Hassan; White, Kenneth
Membrane proteins (MPs) are large set of biological macromolecules that play a fundamental role in physiology and pathophysiology for survival. From a pharma-economical perspective, though it is the fact that MPs constitute ˜75% of possible targets for novel drugs but MPs are one of the most understudied groups of proteins in biochemical research. This is mainly because of the technical difficulties of obtaining structural information about trans-membrane regions (these are small sequences that crossways the bilayer lipid membrane). It is quite useful to predict the location of transmembrane segments down the sequence, since these are the elementary structural building blocks defining their topology. There have been several attempts over the last 20 years to develop tools for predicting membrane-spanning regions but current tools are far away from achieving a considerable reliability in prediction. This study aims to exploit the knowledge and current understanding in the field of artificial neural networks (ANNs) in particular data representation through the development of a system to identify and predict membrane-spanning regions by analysing primary amino acids sequence. In this paper we present a novel neural network (NNs) architecture and algorithms for predicting membrane spanning regions from primary amino acids sequences by using their preference parameters.
Directory of Open Access Journals (Sweden)
Ahmadi Majid
2003-01-01
Full Text Available This paper introduces a novel method for the recognition of human faces in digital images using a new feature extraction method that combines the global and local information in frontal view of facial images. Radial basis function (RBF neural network with a hybrid learning algorithm (HLA has been used as a classifier. The proposed feature extraction method includes human face localization derived from the shape information. An efficient distance measure as facial candidate threshold (FCT is defined to distinguish between face and nonface images. Pseudo-Zernike moment invariant (PZMI with an efficient method for selecting moment order has been used. A newly defined parameter named axis correction ratio (ACR of images for disregarding irrelevant information of face images is introduced. In this paper, the effect of these parameters in disregarding irrelevant information in recognition rate improvement is studied. Also we evaluate the effect of orders of PZMI in recognition rate of the proposed technique as well as RBF neural network learning speed. Simulation results on the face database of Olivetti Research Laboratory (ORL indicate that the proposed method for human face recognition yielded a recognition rate of 99.3%.
Lobato, Justo; Cañizares, Pablo; Rodrigo, Manuel A.; Linares, José J.; Piuleac, Ciprian-George; Curteanu, Silvia
Neural network models represent an important tool of Artificial Intelligence for fuel cell researchers in order to help them to elucidate the processes within the cells, by allowing optimization of materials, cells, stacks, and systems and support control systems. In this work three types of neural networks, that have as common characteristic the supervised learning control (Multilayer Perceptron, Generalized Feedforward Network and Jordan and Elman Network), have been designed to model the performance of a polybenzimidazole-polymer electrolyte membrane fuel cells operating upon a temperature range of 100-175 °C. The influence of temperature of two periods was studied: the temperature in the conditioning period and temperature when the fuel cell was operating. Three inputs variables: the conditioning temperature, the operating temperature and current density were taken into account in order to evaluate their influence upon the potential, the cathode resistance and the ohmic resistance. The Multilayer Perceptron model provides good predictions for different values of operating temperatures and potential and, hence, it is the best choice among the study models, recommended to investigate the influence of process variables of PEMFCs.
Parameter estimation of an aeroelastic aircraft using neural networks
Indian Academy of Sciences (India)
Application of neural networks to the problem of aerodynamic modelling and parameter estimation for aeroelastic aircraft is addressed. A neural model capable of predicting generalized force and moment coefficients using measured motion and control variables only, without any need for conventional normal elastic ...
Parameter estimation of an aeroelastic aircraft using neural networks
Indian Academy of Sciences (India)
e-mail: scr@iitk.ac.in. Abstract. Application of neural networks to the problem of aerodynamic modelling and parameter estimation for aeroelastic aircraft is addressed. A neural model capable of ... of the network in terms of the number of neurons in the hidden layer, the learning rate, the momentum rate etc. is not an exact ...
De, Suvranu; Deo, Dhannanjay; Sankaranarayanan, Ganesh; Arikatla, Venkata S
2011-08-01
BACKGROUND: While an update rate of 30 Hz is considered adequate for real time graphics, a much higher update rate of about 1 kHz is necessary for haptics. Physics-based modeling of deformable objects, especially when large nonlinear deformations and complex nonlinear material properties are involved, at these very high rates is one of the most challenging tasks in the development of real time simulation systems. While some specialized solutions exist, there is no general solution for arbitrary nonlinearities. METHODS: In this work we present PhyNNeSS - a Physics-driven Neural Networks-based Simulation System - to address this long-standing technical challenge. The first step is an off-line pre-computation step in which a database is generated by applying carefully prescribed displacements to each node of the finite element models of the deformable objects. In the next step, the data is condensed into a set of coefficients describing neurons of a Radial Basis Function network (RBFN). During real-time computation, these neural networks are used to reconstruct the deformation fields as well as the interaction forces. RESULTS: We present realistic simulation examples from interactive surgical simulation with real time force feedback. As an example, we have developed a deformable human stomach model and a Penrose-drain model used in the Fundamentals of Laparoscopic Surgery (FLS) training tool box. CONCLUSIONS: A unique computational modeling system has been developed that is capable of simulating the response of nonlinear deformable objects in real time. The method distinguishes itself from previous efforts in that a systematic physics-based pre-computational step allows training of neural networks which may be used in real time simulations. We show, through careful error analysis, that the scheme is scalable, with the accuracy being controlled by the number of neurons used in the simulation. PhyNNeSS has been integrated into SoFMIS (Software Framework for Multimodal
Jafari, Mehdi; Kasaei, Shohreh
2012-01-01
Automatic brain tissue segmentation is a crucial task in diagnosis and treatment of medical images. This paper presents a new algorithm to segment different brain tissues, such as white matter (WM), gray matter (GM), cerebral spinal fluid (CSF), background (BKG), and tumor tissues. The proposed technique uses the modified intraframe coding yielded from H.264/(AVC), for feature extraction. Extracted features are then imposed to an artificial back propagation neural network (BPN) classifier to assign each block to its appropriate class. Since the newest coding standard, H.264/AVC, has the highest compression ratio, it decreases the dimension of extracted features and thus yields to a more accurate classifier with low computational complexity. The performance of the BPN classifier is evaluated using the classification accuracy and computational complexity terms. The results show that the proposed technique is more robust and effective with low computational complexity compared to other recent works.
Döhler, Florian; Mormann, Florian; Weber, Bernd; Elger, Christian E; Lehnertz, Klaus
2008-05-30
We present a cellular neuronal network (CNN) based approach to classify magnetic resonance images with and without hippocampal or Ammon's horn sclerosis (AHS) in the medial temporal lobe. A CNN combines the architecture of cellular automata and artificial neural networks and is an array of locally coupled nonlinear electrical circuits or cells, which is capable of processing a large amount of information in parallel and in real time. Using an exemplary database that consists of a large number of volumes of interest extracted from T1-weighted magnetic resonance images from 144 subjects we here demonstrate that the network allows to classify brain tissue with respect to the presence or absence of mesial temporal sclerosis. Results indicate the general feasibility of CNN-based computer-aided systems for diagnosis and classification of images generated by medical imaging systems.
Shahlaei, M; Fassihi, A; Saghaie, L; Arkan, E; Pourhossein, A
2011-01-01
A quantitative structure activity relationship (QSAR) model based on artificial neural networks (ANN) was developed to study the activities of 29 derivatives of 3-amino-4-(2-(2-(4-benzylpiperazin-1-yl)-2-oxoethoxy) phenylamino) cyclobutenedione as C-C chemokine receptor type 1(CCR1) inhibitors. A feed-forward ANN with error back-propagation learning algorithm was used for model building which was achieved by optimizing initial learning rate, learning momentum, epoch and the number of hidden neurons. Good results were obtained with a Root Mean Square Error (RMSE) and correlation coefficients (R(2)) of 0.189 and 0.906 for the training and 0.103 and 0.932 prediction sets, respectively. The results reflect a nonlinear relationship between the Principal components obtained from calculated molecular descriptors and the inhibitory activities of the investigated molecules.
Directory of Open Access Journals (Sweden)
E Arkan
2011-12-01
Full Text Available Background and the purpose of the study: A quantitative structure activity relationship (QSAR model based on artificial neural networks (ANN was developed to study the activities of 29 derivatives of 3-amino-4-(2-(2-(4-benzylpiperazin-1-yl-2-oxoethoxy phenylamino cyclobutenedione as C-C chemokine receptor type 1(CCR1 inhibitors. Methods: A feed-forward ANN with error back-propagation learning algorithm was used for model building which was achieved by optimizing initial learning rate, learning momentum, epoch and the number of hidden neurons. Results: Good results were obtained with a Root Mean Square Error (RMSE and correlation coefficients (R2 of 0.189 and 0.906 for the training and 0.103 and 0.932 prediction sets, respectively. Conclusion: The results reflect a nonlinear relationship between the Principal components obtained from calculated molecular descriptors and the inhibitory activities of the investigated molecules.
Directory of Open Access Journals (Sweden)
WenBo Xiao
Full Text Available In this article, we introduced an artificial neural network (ANN based computational model to predict the output power of three types of photovoltaic cells, mono-crystalline (mono-, multi-crystalline (multi-, and amorphous (amor- crystalline. The prediction results are very close to the experimental data, and were also influenced by numbers of hidden neurons. The order of the solar generation power output influenced by the external conditions from smallest to biggest is: multi-, mono-, and amor- crystalline silicon cells. In addition, the dependences of power prediction on the number of hidden neurons were studied. For multi- and amorphous crystalline cell, three or four hidden layer units resulted in the high correlation coefficient and low MSEs. For mono-crystalline cell, the best results were achieved at the hidden layer unit of 8.
Xiao, WenBo; Nazario, Gina; Wu, HuaMing; Zhang, HuaMing; Cheng, Feng
2017-01-01
In this article, we introduced an artificial neural network (ANN) based computational model to predict the output power of three types of photovoltaic cells, mono-crystalline (mono-), multi-crystalline (multi-), and amorphous (amor-) crystalline. The prediction results are very close to the experimental data, and were also influenced by numbers of hidden neurons. The order of the solar generation power output influenced by the external conditions from smallest to biggest is: multi-, mono-, and amor- crystalline silicon cells. In addition, the dependences of power prediction on the number of hidden neurons were studied. For multi- and amorphous crystalline cell, three or four hidden layer units resulted in the high correlation coefficient and low MSEs. For mono-crystalline cell, the best results were achieved at the hidden layer unit of 8.
Neural networks for estimation of ocean wave parameters
Digital Repository Service at National Institute of Oceanography (India)
Mandal, S.; Rao, S.; Raju, D.H.
Ocean wave parameters play a significant role in the design of all coastal and offshore structures. In the present study, neural networks are used to estimate various ocean wave parameters from theoretical Pierson-Moskowitz spectra as well...
Pavone, Andrea; Svensson, Jakob; Langenberg, Andreas; Pablant, Novimir; Wolf, Robert C.
2017-10-01
Artificial neural networks (ANNs) can reduce the computation time required for the application of Bayesian inference on large amounts of data by several orders of magnitude, making real-time analysis possible and, at the same time, providing a reliable alternative to more conventional inversion routines. The large scale fusion experiment Wendelstein 7-X (W7-X) requires tens of diagnostics for plasma parameter measurements and is using the Minerva Bayesian modelling framework as its main inference engine, which can handle joint inference in complex systems made of several physics models. Conventional inversion routines are applied to measured data to infer the posterior distribution of the free parameters of the models implemented in the framework. We have trained ANNs on a training set made of samples from the prior distribution of the free parameters and the corresponding data calculated with the forward model, so that the trained ANNs constitute a surrogate model of the physics model. The ANNs have been then applied to 2D images measured by an X-ray spectrometer, representing the spectral emission from plasma impurities measured along a fan of lines of sight covering a major fraction of the plasma cross-section, for the inference of ion temperature profiles and then compared with the conventional inversion routines, showing that they constitute a robust and reliable alternative for real time plasma parameter inference.
Directory of Open Access Journals (Sweden)
S. Saravanan
2012-07-01
Full Text Available Power System planning starts with Electric load (demand forecasting. Accurate electricity load forecasting is one of the most important challenges in managing supply and demand of the electricity, since the electricity demand is volatile in nature; it cannot be stored and has to be consumed instantly. The aim of this study deals with electricity consumption in India, to forecast future projection of demand for a period of 19 years from 2012 to 2030. The eleven input variables used are Amount of CO2 emission, Population, Per capita GDP, Per capita gross national income, Gross Domestic savings, Industry, Consumer price index, Wholesale price index, Imports, Exports and Per capita power consumption. A new methodology based on Artificial Neural Networks (ANNs using principal components is also used. Data of 29 years used for training and data of 10 years used for testing the ANNs. Comparison made with multiple linear regression (based on original data and the principal components and ANNs with original data as input variables. The results show that the use of ANNs with principal components (PC is more effective.
Lemke, Tobias; Peter, Christine
2017-12-12
Coarse-grained (CG) simulation models have become very popular tools to study complex molecular systems with great computational efficiency on length and time scales that are inaccessible to simulations at atomistic resolution. In so-called bottom-up coarse-graining strategies, the interactions in the CG model are devised such that an accurate representation of an atomistic sampling of configurational phase space is achieved. This means the coarse-graining methods use the underlying multibody potential of mean force (i.e., free-energy surface) derived from the atomistic simulation as parametrization target. Here, we present a new method where a neural network (NN) is used to extract high-dimensional free energy surfaces (FES) from molecular dynamics (MD) simulation trajectories. These FES are used for simulations on a CG level of resolution. The method is applied to simulating homo-oligo-peptides (oligo-glutamic-acid (oligo-glu) and oligo-aspartic-acid (oligo-asp)) of different lengths. We show that the NN not only is able to correctly describe the free-energy surface for oligomer lengths that it was trained on but also is able to predict the conformational sampling of longer chains.
Energy Technology Data Exchange (ETDEWEB)
Piggott, W T; Aceves, S M; Flowers, D L; Chen, J Y
2007-09-26
We have added the capability to look at in-cylinder fuel distributions using a previously developed ignition model within a fluid mechanics code (KIVA3V) that uses an artificial neural network (ANN) to predict ignition (The combined code: KIVA3V-ANN). KIVA3V-ANN was originally developed and validated for analysis of Homogeneous Charge Compression Ignition (HCCI) combustion, but it is also applicable to the more difficult problem of Premixed Charge Compression Ignition (PCCI) combustion. PCCI combustion refers to cases where combustion occurs as a nonmixing controlled, chemical kinetics dominated, autoignition process, where the fuel, air, and residual gas mixtures are not necessarily as homogeneous as in HCCI combustion. This paper analyzes the effects of introducing charge non-uniformity into a KIVA3V-ANN simulation. The results are compared to experimental results, as well as simulation results using a more physically representative and computationally intensive code (KIVA3V-MPI-MZ), which links a fluid mechanics code to a multi-zone detailed chemical kinetics solver. The results indicate that KIVA3V-ANN produces reasonable approximations to the more accurate KIVA3V-MPI-MZ at a much reduced computational cost.
Heddam, Salim
2016-09-01
This paper proposes multilayer perceptron neural network (MLPNN) to predict phycocyanin (PC) pigment using water quality variables as predictor. In the proposed model, four water quality variables that are water temperature, dissolved oxygen, pH, and specific conductance were selected as the inputs for the MLPNN model, and the PC as the output. To demonstrate the capability and the usefulness of the MLPNN model, a total of 15,849 data measured at 15-min (15 min) intervals of time are used for the development of the model. The data are collected at the lower Charles River buoy, and available from the US Environmental Protection Agency (USEPA). For comparison purposes, a multiple linear regression (MLR) model that was frequently used for predicting water quality variables in previous studies is also built. The performances of the models are evaluated using a set of widely used statistical indices. The performance of the MLPNN and MLR models is compared with the measured data. The obtained results show that (i) the all proposed MLPNN models are more accurate than the MLR models and (ii) the results obtained are very promising and encouraging for the development of phycocyanin-predictive models.
Shukla, Pitamber; Basu, Ishita; Graupe, Daniel; Tuninetti, Daniela; Slavin, Konstantin V
2012-01-01
The current Food and Drug Administration approved system for the treatment of tremor disorders through Deep Brain Stimulation (DBS) of the area of the brain that controls movement, operates open-loop. It does not automatically adapt to the instantaneous patient's needs or to the progression of the disease. This paper demonstrates an adaptive closed-loop controlled DBS that, after switching off stimulation, tracks few physiological signals to predict the reappearance of tremor before the patient experiences discomfort, at which point it instructs the DBS controller to switch on stimulation again. The core of the proposed approach is a Neural Network (NN) which effectively extracts tremor predictive information from non-invasively recorded surface-electromyogram(sEMG) and accelerometer signals measured at the symptomatic extremities. A simple feed-forward back-propagation NN architecture is shown to successfully predict tremor in 31 out of 33 trials in two Parkinson's Disease patients with an overall accuracy of 75.8% and sensitivity of 92.3%. This work therefore shows that closed-loop DBS control is feasible in the near future and that it can be achieved without modifications of the electrodes implanted in the brain, i.e., is backward compatible with approved DBS systems.
Directory of Open Access Journals (Sweden)
Bin Pan
2017-01-01
Full Text Available Aircraft detection from high-resolution remote sensing images is important for civil and military applications. Recently, detection methods based on deep learning have rapidly advanced. However, they require numerous samples to train the detection model and cannot be directly used to efficiently handle large-area remote sensing images. A weakly supervised learning method (WSLM can detect a target with few samples. However, it cannot extract an adequate number of features, and the detection accuracy requires improvement. We propose a cascade convolutional neural network (CCNN framework based on transfer-learning and geometric feature constraints (GFC for aircraft detection. It achieves high accuracy and efficient detection with relatively few samples. A high-accuracy detection model is first obtained using transfer-learning to fine-tune pretrained models with few samples. Then, a GFC region proposal filtering method improves detection efficiency. The CCNN framework completes the aircraft detection for large-area remote sensing images. The framework first-level network is an image classifier, which filters the entire image, excluding most areas with no aircraft. The second-level network is an object detector, which rapidly detects aircraft from the first-level network output. Compared with WSLM, detection accuracy increased by 3.66%, false detection decreased by 64%, and missed detection decreased by 23.1%.
Directory of Open Access Journals (Sweden)
Yiming Jiang
2016-01-01
Full Text Available Over the last few decades, the intelligent control methods such as fuzzy logic control (FLC and neural network (NN control have been successfully used in various applications. The rapid development of digital computer based control systems requires control signals to be calculated in a digital or discrete-time form. In this background, the intelligent control methods developed for discrete-time systems have drawn great attentions. This survey aims to present a summary of the state of the art of the design of FLC and NN-based intelligent control for discrete-time systems. For discrete-time FLC systems, numerous remarkable design approaches are introduced and a series of efficient methods to deal with the robustness, stability, and time delay of FLC discrete-time systems are recommended. Techniques for NN-based intelligent control for discrete-time systems, such as adaptive methods and adaptive dynamic programming approaches, are also reviewed. Overall, this paper is devoted to make a brief summary for recent progresses in FLC and NN-based intelligent control design for discrete-time systems as well as to present our thoughts and considerations of recent trends and potential research directions in this area.
Parameter estimation using compensatory neural networks
Indian Academy of Sciences (India)
Proposed here is a new neuron model, a basis for Compensatory Neural Network Architecture (CNNA), which not only reduces the total number of interconnections among neurons but also reduces the total computing time for training. The suggested model has properties of the basic neuron model as well as the higher ...
Yaşar, Hüseyin; Ceylan, Murat
2015-03-01
Breast cancer is one of the types of cancer which is most commonly seen in women. Density of breast is an important indicator for the risk of cancer. In addition, densities of tissue may harden the diagnosis by hiding the abnormalities occurring on the breast. For this reason, during the process of diagnosis, the process of automatic classification of breast density has a significant importance. In this study, a new system with the base of Artificial Neural Network (ANN) and multiple resolution analysis is suggested. Wavelet and curvelet analyses having the most common use have been used as multi resolution analysis. 4 pieces of statistics which are minimum value, maximum value, mean value and standard deviation have been extracted from the images which have been eluted to their sub-bands via multi resolution analysis. For the purpose of testing the success of the system, 322 pieces of images which are in MIAS database have been used. The obtained results for different backgrounds are so satisfying; and the highest classification values have been obtained as 97.16 % with Wavelet transform and ANN for fatty background and 79.80 % with Wavelet transform and ANN for fatty-glanduar background. The same results have been obtained using Wavelet transform and ANN and Curvelet transform and ANN for dense background and accuracy rate of 84.82 % have been reached. The results of mean classification have been obtained, for three pieces of tissue types (fatty, fatty-glanduar, dense), in sequence as 84.47 % with the use of ANN, 85.71 % with the use of curvelet analysis and ANN; and 87.26 % with the use of wavelet analysis and ANN.
Directory of Open Access Journals (Sweden)
Khastavaneh H.
2017-06-01
Full Text Available Background: Multiple Sclerosis (MS is a degenerative disease of central nervous system. MS patients have some dead tissues in their brains called MS lesions. MRI is an imaging technique sensitive to soft tissues such as brain that shows MS lesions as hyper-intense or hypo-intense signals. Since manual segmentation of these lesions is a laborious and time consuming task, automatic segmentation is a need. Materials and Methods: In order to segment MS lesions, a method based on learning kernels has been proposed. The proposed method has three main steps namely; pre-processing, sub-region extraction and segmentation. The segmentation is performed by a kernel. This kernel is trained using a modified version of a special type of Artificial Neural Networks (ANN called Massive Training ANN (MTANN. The kernel incorporates surrounding pixel information as features for classification of middle pixel of kernel. The materials of this study include a part of MICCAI 2008 MS lesion segmentation grand challenge data-set. Results: Both qualitative and quantitative results show promising results. Similarity index of 70 percent in some cases is considered convincing. These results are obtained from information of only one MRI channel rather than multi-channel MRIs. Conclusion: This study shows the potential of surrounding pixel information to be incorporated in segmentation by learning kernels. The performance of proposed method will be improved using a special pre-processing pipeline and also a post-processing step for reducing false positives/negatives. An important advantage of proposed model is that it uses just FLAIR MRI that reduces computational time and brings comfort to patients.
Khastavaneh, H; Ebrahimpour-Komleh, H
2017-06-01
Multiple Sclerosis (MS) is a degenerative disease of central nervous system. MS patients have some dead tissues in their brains called MS lesions. MRI is an imaging technique sensitive to soft tissues such as brain that shows MS lesions as hyper-intense or hypo-intense signals. Since manual segmentation of these lesions is a laborious and time consuming task, automatic segmentation is a need. In order to segment MS lesions, a method based on learning kernels has been proposed. The proposed method has three main steps namely; pre-processing, sub-region extraction and segmentation. The segmentation is performed by a kernel. This kernel is trained using a modified version of a special type of Artificial Neural Networks (ANN) called Massive Training ANN (MTANN). The kernel incorporates surrounding pixel information as features for classification of middle pixel of kernel. The materials of this study include a part of MICCAI 2008 MS lesion segmentation grand challenge data-set. Both qualitative and quantitative results show promising results. Similarity index of 70 percent in some cases is considered convincing. These results are obtained from information of only one MRI channel rather than multi-channel MRIs. This study shows the potential of surrounding pixel information to be incorporated in segmentation by learning kernels. The performance of proposed method will be improved using a special pre-processing pipeline and also a post-processing step for reducing false positives/negatives. An important advantage of proposed model is that it uses just FLAIR MRI that reduces computational time and brings comfort to patients.
Sesin, Anaelis; Adjouadi, Malek; Cabrerizo, Mercedes; Ayala, Melvin; Barreto, Armando
2008-01-01
This study developed an adaptive real-time human-computer interface (HCI) that serves as an assistive technology tool for people with severe motor disability. The proposed HCI design uses eye gaze as the primary computer input device. Controlling the mouse cursor with raw eye coordinates results in sporadic motion of the pointer because of the saccadic nature of the eye. Even though eye movements are subtle and completely imperceptible under normal circumstances, they considerably affect the accuracy of an eye-gaze-based HCI. The proposed HCI system is novel because it adapts to each specific user's different and potentially changing jitter characteristics through the configuration and training of an artificial neural network (ANN) that is structured to minimize the mouse jitter. This task is based on feeding the ANN a user's initially recorded eye-gaze behavior through a short training session. The ANN finds the relationship between the gaze coordinates and the mouse cursor position based on the multilayer perceptron model. An embedded graphical interface is used during the training session to generate user profiles that make up these unique ANN configurations. The results with 12 subjects in test 1, which involved following a moving target, showed an average jitter reduction of 35%; the results with 9 subjects in test 2, which involved following the contour of a square object, showed an average jitter reduction of 53%. For both results, the outcomes led to trajectories that were significantly smoother and apt at reaching fixed or moving targets with relative ease and within a 5% error margin or deviation from desired trajectories. The positive effects of such jitter reduction are presented graphically for visual appreciation.
A neural network-based foF2 model for a single station in the polar cap
Athieno, R.; Jayachandran, P. T.; Themens, D. R.
2017-06-01
A neural network (NN) model has been developed for the critical frequency of the F2 layer (foF2) at Resolute (74.70°N, 265.10°E) using data obtained from the Space Physics Interactive Data Resource (no longer available) for the period between 1975 and 1995. This model is a first step toward addressing the discrepancies of the International Reference Ionosphere (IRI) foF2 or peak electron density (NmF2) at high latitudes recently presented by Themens et al. (2014). The performance of the NN model has been evaluated using foF2 data obtained from the Canadian Advanced Digital Ionosonde at Resolute (74.75°N, 265.00°E) for the period between 2009 and 2013, in comparison with the IRI predictions. The 2012 version and the International Union of Radio Science option of IRI have been used. The NN nighttime monthly median foF2 variation demonstrates good agreement with observations compared to the IRI. The NN model is able to reproduce the enhancements in foF2 during the equinoxes and also shows an improvement during disturbed days. Root mean square errors were computed between hourly and monthly median model predictions and observations, and on the whole, the NN model seems to perform better during low solar activity and the equinoxes. The NN model shows an improvement in performance on average by 26.638% for hourly foF2 and 32.636% for monthly median foF2, compared to 7.877% obtained for the same station by the most recent NN model that attempted to predict foF2 at a polar cap station (Oyeyemi and Poole, 2005).
Tomatis, S.; Rancati, T.; Fiorino, C.; Vavassori, V.; Fellin, G.; Cagna, E.; Mauro, F. A.; Girelli, G.; Monti, A.; Baccolini, M.; Naldi, G.; Bianchi, C.; Menegotti, L.; Pasquino, M.; Stasi, M.; Valdagni, R.
2012-03-01
The aim of this study was to develop a model exploiting artificial neural networks (ANNs) to correlate dosimetric and clinical variables with late rectal bleeding in prostate cancer patients undergoing radical radiotherapy and to compare the ANN results with those of a standard logistic regression (LR) analysis. 718 men included in the AIROPROS 0102 trial were analyzed. This multicenter protocol was characterized by the prospective evaluation of rectal toxicity, with a minimum follow-up of 36 months. Radiotherapy doses were between 70 and 80 Gy. Information was recorded for comorbidity, previous abdominal surgery, use of drugs and hormonal therapy. For each patient, a rectal dose-volume histogram (DVH) of the whole treatment was recorded and the equivalent uniform dose (EUD) evaluated as an effective descriptor of the whole DVH. Late rectal bleeding of grade ≥ 2 was considered to define positive events in this study (52 of 718 patients). The overall population was split into training and verification sets, both of which were involved in model instruction, and a test set, used to evaluate the predictive power of the model with independent data. Fourfold cross-validation was also used to provide realistic results for the full dataset. The LR was performed on the same data. Five variables were selected to predict late rectal bleeding: EUD, abdominal surgery, presence of hemorrhoids, use of anticoagulants and androgen deprivation. Following a receiver operating characteristic analysis of the independent test set, the areas under the curves (AUCs) were 0.704 and 0.655 for ANN and LR, respectively. When evaluated with cross-validation, the AUC was 0.714 for ANN and 0.636 for LR, which differed at a significance level of p = 0.03. When a practical discrimination threshold was selected, ANN could classify data with sensitivity and specificity both equal to 68.0%, whereas these values were 61.5% for LR. These data provide reasonable evidence that results obtained with
Agarwal, Smriti; Bisht, Amit Singh; Singh, Dharmendra; Pathak, Nagendra Prasad
2014-12-01
Millimetre wave imaging (MMW) is gaining tremendous interest among researchers, which has potential applications for security check, standoff personal screening, automotive collision-avoidance, and lot more. Current state-of-art imaging techniques viz. microwave and X-ray imaging suffers from lower resolution and harmful ionizing radiation, respectively. In contrast, MMW imaging operates at lower power and is non-ionizing, hence, medically safe. Despite these favourable attributes, MMW imaging encounters various challenges as; still it is very less explored area and lacks suitable imaging methodology for extracting complete target information. Keeping in view of these challenges, a MMW active imaging radar system at 60 GHz was designed for standoff imaging application. A C-scan (horizontal and vertical scanning) methodology was developed that provides cross-range resolution of 8.59 mm. The paper further details a suitable target identification and classification methodology. For identification of regular shape targets: mean-standard deviation based segmentation technique was formulated and further validated using a different target shape. For classification: probability density function based target material discrimination methodology was proposed and further validated on different dataset. Lastly, a novel artificial neural network based scale and rotation invariant, image reconstruction methodology has been proposed to counter the distortions in the image caused due to noise, rotation or scale variations. The designed neural network once trained with sample images, automatically takes care of these deformations and successfully reconstructs the corrected image for the test targets. Techniques developed in this paper are tested and validated using four different regular shapes viz. rectangle, square, triangle and circle.
Cui, Y.; Long, D.; Hong, Y.; Zeng, C.; Han, Z.
2016-12-01
Reconstruction of FY-3B/MWRI soil moisture using an artificial neural network based on reconstructed MODIS optical products over the Tibetan Plateau Yaokui Cui, Di Long, Yang Hong, Chao Zeng, and Zhongying Han State Key Laboratory of Hydroscience and Engineering, Department of Hydraulic Engineering, Tsinghua University, Beijing 100084, China Abstract: Soil moisture is a key variable in the exchange of water and energy between the land surface and the atmosphere, especially over the Tibetan Plateau (TP) which is climatically and hydrologically sensitive as the world's third pole. Large-scale consistent and continuous soil moisture datasets are of importance to meteorological and hydrological applications, such as weather forecasting and drought monitoring. The Fengyun-3B Microwave Radiation Imager (FY-3B/MWRI) soil moisture product is one of relatively new passive microwave products. The FY-3B/MWRI soil moisture product is reconstructed using the back-propagation neural network (BP-NN) based on reconstructed MODIS products, i.e., LST, NDVI, and albedo using different gap-filling methods. The reconstruction method of generating the soil moisture product not only considers the relationship between the soil moisture and the NDVI, LST, and albedo, but also the relationship between the soil moisture and the four-dimensional variation using the longitude, latitude, DEM and day of year (DOY). Results show that the soil moisture could be well reconstructed with R2 larger than 0.63, and RMSE less than 0.1 cm3 cm-3 and bias less than 0.07 cm3 cm-3 for both frozen and unfrozen periods, compared with in-situ measurements in the central TP. The reconstruction method is subsequently applied to generate spatially consistent and temporally continuous surface soil moisture over the TP. The reconstructed FY-3B/MWRI soil moisture product could be valuable in studying meteorology, hydrology, and agriculture over the TP. Keywords: FY-3B/MWRI; Soil moisture; Reconstruction; Tibetan
Flood estimation: a neural network approach
Energy Technology Data Exchange (ETDEWEB)
Swain, P.C.; Seshachalam, C.; Umamahesh, N.V. [Regional Engineering Coll., Warangal (India). Water and Environment Div.
2000-07-01
The artificial neural network (ANN) approach described in this study aims at predicting the flood flow into a reservoir. This differs from the traditional methods of flow prediction in the sense that it belongs to a class of data driven approaches, where as the traditional methods are model driven. Physical processes influencing the occurrences of streamflow in a river are highly complex, and are very difficult to be modelled by available statistical or deterministic models. ANNs provide model free solutions and hence can be expected to be appropriate in these conditions. Non-linearity, adaptivity, evidential response and fault tolerance are additional properties and capabilities of the neural networks. This paper highlights the applicability of neural networks for predicting daily flood flow taking the Hirakud reservoir on river Mahanadi in Orissa, India as the case study. The correlation between the observed and predicted flows and the relative error are considered to measure the performance of the model. The correlation between the observed and the modelled flows are computed to be 0.9467 in testing phase of the model. (orig.)
Kim, Sungho
2017-06-01
Automatic target recognition (ATR) is a traditionally challenging problem in military applications because of the wide range of infrared (IR) image variations and the limited number of training images. IR variations are caused by various three-dimensional target poses, noncooperative weather conditions (fog and rain), and difficult target acquisition environments. Recently, deep convolutional neural network-based approaches for RGB images (RGB-CNN) showed breakthrough performance in computer vision problems, such as object detection and classification. The direct use of RGB-CNN to the IR ATR problem fails to work because of the IR database problems (limited database size and IR image variations). An IR variation-reduced deep CNN (IVR-CNN) to cope with the problems is presented. The problem of limited IR database size is solved by a commercial thermal simulator (OKTAL-SE). The second problem of IR variations is mitigated by the proposed shifted ramp function-based intensity transformation. This can suppress the background and enhance the target contrast simultaneously. The experimental results on the synthesized IR images generated by the thermal simulator (OKTAL-SE) validated the feasibility of IVR-CNN for military ATR applications.
Directory of Open Access Journals (Sweden)
J. Uwamahoro
2012-06-01
Full Text Available Estimating the geoeffectiveness of solar events is of significant importance for space weather modelling and prediction. This paper describes the development of a neural network-based model for estimating the probability occurrence of geomagnetic storms following halo coronal mass ejection (CME and related interplanetary (IP events. This model incorporates both solar and IP variable inputs that characterize geoeffective halo CMEs. Solar inputs include numeric values of the halo CME angular width (AW, the CME speed (Vcme, and the comprehensive flare index (cfi, which represents the flaring activity associated with halo CMEs. IP parameters used as inputs are the numeric peak values of the solar wind speed (Vsw and the southward Z-component of the interplanetary magnetic field (IMF or Bs. IP inputs were considered within a 5-day time window after a halo CME eruption. The neural network (NN model training and testing data sets were constructed based on 1202 halo CMEs (both full and partial halo and their properties observed between 1997 and 2006. The performance of the developed NN model was tested using a validation data set (not part of the training data set covering the years 2000 and 2005. Under the condition of halo CME occurrence, this model could capture 100% of the subsequent intense geomagnetic storms (Dst ≤ −100 nT. For moderate storms (−100 < Dst ≤ −50, the model is successful up to 75%. This model's estimate of the storm occurrence rate from halo CMEs is estimated at a probability of 86%.
Directory of Open Access Journals (Sweden)
Erna Rusliana Muhamad Saleh
2014-02-01
Full Text Available Wafer is type of biscuit frequently found on expired condition in market, therefore prediction method should be implemented to avoid this condition. apart from the prediction of shelf-life of wafer done by laboratory test, which were time-consuming, expensive, required trained panelists, complex equipment and suitable ambience, artificial neural network (ANN based dielectric parameters was proposed in nthis study. The aim of study was to develop model to predict shelf-life employing aNN based capacitance parameter. Back propagation algorithm with trial and error was applied in variations of nodes per hidden layer, number of hidden layers, activation functions, the function of learnings and epochs. The result of study was the model was able to predict wafer shelf-life. The accuracy level was shown by low MSE value (0.01 and high coefficient correlation value (89.25%. Keywords: artificial Neural Network, shelf-life, waffer, dielectric, capacitance ABSTRAK Wafer adalah jenis makanan kering yang sering ditemukan kedaluwarsa. Penentuan masa kedaluwarsa dengan observasi laboratorium memiliki beberapa kelemahan, diantaranya memakan waktu, panelis terlatih, suasana yang tepat, biaya dan alat uji yang kompleks. alternatif solusinya adalah penggunaan artificial Neural Network (ANN berbasiskan parameter kapasitansi. Tujuan kerja ilmiah ini adalah untuk memprediksi masa kedaluwarsa wafer menggunakan aNN berbasiskan parameter kapasitansi. algoritma pembelajaran yang digunakan adalah Backpropagation dengan trial and error variasi jumlah node per hidden layer, jumlah hidden layer, fungsi aktivasi, fungsi pembelajaran dan epoch. Hasil prediksi menunjukkan bahwa aNN hasil pelatihan yang dikombinasikan dengan parameter kapasitansi mampu memprediksi masa kedaluwarsa wafer dengan MSE terendah 0,01 dan R tertinggi 89,25%. Kata kunci: Jaringan Syaraf Tiruan, masa kedaluwarsa, wafer, dielektrik, kapasitansi
Optimizing Neural Network Architectures Using Generalization Error Estimators
DEFF Research Database (Denmark)
Larsen, Jan
1994-01-01
This paper addresses the optimization of neural network architectures. It is suggested to optimize the architecture by selecting the model with minimal estimated averaged generalization error. We consider a least-squares (LS) criterion for estimating neural network models, i.e., the associated...... model weights are estimated by minimizing the LS criterion. The quality of a particular estimated model is measured by the average generalization error. This is defined as the expected squared prediction error on a novel input-output sample averaged over all possible training sets. An essential part...... of the suggested architecture optimization scheme is to calculate an estimate of the average generalization error. We suggest using the GEN-estimator which allows for dealing with nonlinear, incomplete models, i.e., models which are not capable of modeling the underlying nonlinear relationship perfectly. In most...
Fuzzy/Neural Software Estimates Costs of Rocket-Engine Tests
Douglas, Freddie; Bourgeois, Edit Kaminsky
2005-01-01
The Highly Accurate Cost Estimating Model (HACEM) is a software system for estimating the costs of testing rocket engines and components at Stennis Space Center. HACEM is built on a foundation of adaptive-network-based fuzzy inference systems (ANFIS) a hybrid software concept that combines the adaptive capabilities of neural networks with the ease of development and additional benefits of fuzzy-logic-based systems. In ANFIS, fuzzy inference systems are trained by use of neural networks. HACEM includes selectable subsystems that utilize various numbers and types of inputs, various numbers of fuzzy membership functions, and various input-preprocessing techniques. The inputs to HACEM are parameters of specific tests or series of tests. These parameters include test type (component or engine test), number and duration of tests, and thrust level(s) (in the case of engine tests). The ANFIS in HACEM are trained by use of sets of these parameters, along with costs of past tests. Thereafter, the user feeds HACEM a simple input text file that contains the parameters of a planned test or series of tests, the user selects the desired HACEM subsystem, and the subsystem processes the parameters into an estimate of cost(s).
Vehicle Sideslip Angle Estimation Based on General Regression Neural Network
Directory of Open Access Journals (Sweden)
Wang Wei
2016-01-01
Full Text Available Aiming at the accuracy of estimation of vehicle’s mass center sideslip angle, an estimation method of slip angle based on general regression neural network (GRNN and driver-vehicle closed-loop system has been proposed: regarding vehicle’s sideslip angle as time series mapping of yaw speed and lateral acceleration; using homogeneous design project to optimize the training samples; building the mapping relationship among sideslip angle, yaw speed, and lateral acceleration; at the same time, using experimental method to measure vehicle’s sideslip angle to verify validity of this method. Estimation results of neural network and real vehicle experiment show the same changing tendency. The mean of error is within 10% of test result’s amplitude. Results show GRNN can estimate vehicle’s sideslip angle correctly. It can offer a reference to the application of vehicle’s stability control system on vehicle’s state estimation.
Wu, Jian; Murphy, Martin J
2010-11-01
To develop a neural network based registration quality evaluator (RQE) that can identify unsuccessful 3D/3D image registrations for the head-and-neck patient setup in radiotherapy. A two-layer feed-forward neural network was used as a RQE to classify 3D/3D rigid registration solutions as successful or unsuccessful based on the features of the similarity surface near the point-of-solution. The supervised training and test data sets were generated by rigidly registering daily cone-beam CTs to the treatment planning fan-beam CTs of six patients with head-and-neck tumors. Two different similarity metrics (mutual information and mean-squared intensity difference) and two different types of image content (entire image versus bony landmarks) were used. The best solution for each registration pair was selected from 50 optimizing attempts that differed only by the initial transformation parameters. The distance from each individual solution to the best solution in the normalized parametrical space was compared to a user-defined error threshold to determine whether that solution was successful or not. The supervised training was then used to train the RQE. The performance of the RQE was evaluated using the test data set that consisted of registration results that were not used in training. The RQE constructed using the mutual information had very good performance when tested using the test data sets, yielding the sensitivity, the specificity, the positive predictive value, and the negative predictive value in the ranges of 0.960-1.000, 0.993-1.000, 0.983-1.000, and 0.909-1.000, respectively. Adding a RQE into a conventional 3D/3D image registration system incurs only about 10%-20% increase of the overall processing time. The authors' patient study has demonstrated very good performance of the proposed RQE when used with the mutual information in identifying unsuccessful 3D/3D registrations for daily patient setup. The classifier had very good generality and required only to
Bayesian auxiliary particle filters for estimating neural tuning parameters.
Mountney, John; Sobel, Marc; Obeid, Iyad
2009-01-01
A common challenge in neural engineering is to track the dynamic parameters of neural tuning functions. This work introduces the application of Bayesian auxiliary particle filters for this purpose. Based on Monte-Carlo filtering, Bayesian auxiliary particle filters use adaptive methods to model the prior densities of the state parameters being tracked. The observations used are the neural firing times, modeled here as a Poisson process, and the biological driving signal. The Bayesian auxiliary particle filter was evaluated by simultaneously tracking the three parameters of a hippocampal place cell and compared to a stochastic state point process filter. It is shown that Bayesian auxiliary particle filters are substantially more accurate and robust than alternative methods of state parameter estimation. The effects of time-averaging on parameter estimation are also evaluated.
Voltage Estimation in Active Distribution Grids Using Neural Networks
DEFF Research Database (Denmark)
Pertl, Michael; Heussen, Kai; Gehrke, Oliver
2016-01-01
the observability of distribution systems has to be improved. To increase the situational awareness of the power system operator data driven methods can be employed. These methods benefit from newly available data sources such as smart meters. This paper presents a voltage estimation method based on neural networks...
An Improved Convolutional Neural Network on Crowd Density Estimation
Directory of Open Access Journals (Sweden)
Pan Shao-Yun
2016-01-01
Full Text Available In this paper, a new method is proposed for crowd density estimation. An improved convolutional neural network is combined with traditional texture feature. The data calculated by the convolutional layer can be treated as a new kind of features.So more useful information of images can be extracted by different features.In the meantime, the size of image has little effect on the result of convolutional neural network. Experimental results indicate that our scheme has adequate performance to allow for its use in real world applications.
A neural network applied to estimate Burr XII distribution parameters
Energy Technology Data Exchange (ETDEWEB)
Abbasi, B., E-mail: b.abbasi@gmail.co [Department of Industrial Engineering, Sharif University of Technology, Tehran (Iran, Islamic Republic of); Hosseinifard, S.Z. [Department of Statistics and Operations Research, RMIT University, Melbourne (Australia); Coit, D.W. [Department of Industrial and System Engineering, Rutgers University, Piscataway, NJ (United States)
2010-06-15
The Burr XII distribution can closely approximate many other well-known probability density functions such as the normal, gamma, lognormal, exponential distributions as well as Pearson type I, II, V, VII, IX, X, XII families of distributions. Considering a wide range of shape and scale parameters of the Burr XII distribution, it can have an important role in reliability modeling, risk analysis and process capability estimation. However, estimating parameters of the Burr XII distribution can be a complicated task and the use of conventional methods such as maximum likelihood estimation (MLE) and moment method (MM) is not straightforward. Some tables to estimate Burr XII parameters have been provided by Burr (1942) but they are not adequate for many purposes or data sets. Burr tables contain specific values of skewness and kurtosis and their corresponding Burr XII parameters. Using interpolation or extrapolation to estimate other values may provide inappropriate estimations. In this paper, we present a neural network to estimate Burr XII parameters for different values of skewness and kurtosis as inputs. A trained network is presented, and one can use it without previous knowledge about neural networks to estimate Burr XII distribution parameters. Accurate estimation of the Burr parameters is an extension of simulation studies.
Estimating Ads’ Click through Rate with Recurrent Neural Network
Directory of Open Access Journals (Sweden)
Chen Qiao-Hong
2016-01-01
Full Text Available With the development of the Internet, online advertising spreads across every corner of the world, the ads' click through rate (CTR estimation is an important method to improve the online advertising revenue. Compared with the linear model, the nonlinear models can study much more complex relationships between a large number of nonlinear characteristics, so as to improve the accuracy of the estimation of the ads’ CTR. The recurrent neural network (RNN based on Long-Short Term Memory (LSTM is an improved model of the feedback neural network with ring structure. The model overcomes the problem of the gradient of the general RNN. Experiments show that the RNN based on LSTM exceeds the linear models, and it can effectively improve the estimation effect of the ads’ click through rate.
A spatial neural fuzzy network for estimating pan evaporation at ungauged sites
Directory of Open Access Journals (Sweden)
C.-H. Chung
2012-01-01
Full Text Available Evaporation is an essential reference to the management of water resources. In this study, a hybrid model that integrates a spatial neural fuzzy network with the kringing method is developed to estimate pan evaporation at ungauged sites. The adaptive network-based fuzzy inference system (ANFIS can extract the nonlinear relationship of observations, while kriging is an excellent geostatistical interpolator. Three-year daily data collected from nineteen meteorological stations covering the whole of Taiwan are used to train and test the constructed model. The pan evaporation (E_{pan} at ungauged sites can be obtained through summing up the outputs of the spatially weighted ANFIS and the residuals adjusted by kriging. Results indicate that the proposed AK model (hybriding ANFIS and kriging can effectively improve the accuracy of E_{pan} estimation as compared with that of empirical formula. This hybrid model demonstrates its reliability in estimating the spatial distribution of E_{pan} and consequently provides precise E_{pan} estimation by taking geographical features into consideration.
Pulvirenti, Luca; Pierdicca, Nazzareno; Marzano, Frank S
2008-12-03
A simulation study to assess the potentiality of sea surface wind vector estimation based on the approximation of the forward model through Neural Networks and on the Bayesian theory of parameter estimation is presented. A polarimetric microwave radiometer has been considered and its observations have been simulated by means of the two scale model. To perform the simulations, the atmospheric and surface parameters have been derived from ECMWF analysis fields. To retrieve wind speed, Minimum Variance (MV) and Maximum Posterior Probability (MAP) criteria have been used while, for wind direction, a Maximum Likelihood (ML) criterion has been exploited. To minimize the cost function of MAP and ML, conventional Gradient Descent method, as well as Simulated Annealing optimization technique, have been employed. Results have shown that the standard deviation of the wind speed retrieval error is approximately 1.1 m/s for the best estimator. As for the wind direction, the standard deviation of the estimation error is less than 13° for wind speeds larger than 6 m/s. For lower wind velocities, the wind direction signal is too weak to ensure reliable retrievals. A method to deal with the non-uniqueness of the wind direction solution has been also developed. A test on a case study has yielded encouraging results.
Age estimation of facial image based on convolution neural network
Meng, Xiaodong; Wang, Yifeng; Zheng, Haihong
2017-07-01
Age is an inherent biological characteristic of human and is reflected in facial images to a certain extent. A method for estimating age from a facial image by combining CNN (Convolution Neural Network) with SVR (Support Vector Regression) is proposed. First, a deep CNN is trained to automatically extract age features from facial images and classify them into variant age groups. Then different SVRs are trained for each age group to estimate the age of a facial image. The experimental results show that a lower MAE (Mean Absolute Error) of age estimation on MORPH database is obtained.
Estimating Neural Signal Dynamics in the Human Brain
Directory of Open Access Journals (Sweden)
Christopher W Tyler
2011-06-01
Full Text Available Although brain imaging methods are highly effective for localizing the effects of neural activation throughout the human brain in terms of the blood oxygenation level dependent (BOLD response, there is currently no way to estimate the underlying neural signal dynamics in generating the BOLD response in each local activation region (except for processes slower than the BOLD time course. Knowledge of the neural signal is critical information if spatial mapping is to progress to the analysis of dynamic information flow through the cortical networks as the brain performs its tasks. We introduce an analytic approach that provides a new level of conceptualization and specificity in the study of brain processing by noninvasive methods. This technique allows us to use brain imaging methods to determine the dynamics of local neural population responses to their native temporal resolution throughout the human brain, with relatively narrow confidence intervals on many response properties. The ability to characterize local neural dynamics in the human brain represents a significant enhancement of brain imaging capabilities, with potential application from general cognitive studies to assessment of neuropathologies.
Neural Net Gains Estimation Based on an Equivalent Model
Directory of Open Access Journals (Sweden)
Karen Alicia Aguilar Cruz
2016-01-01
Full Text Available A model of an Equivalent Artificial Neural Net (EANN describes the gains set, viewed as parameters in a layer, and this consideration is a reproducible process, applicable to a neuron in a neural net (NN. The EANN helps to estimate the NN gains or parameters, so we propose two methods to determine them. The first considers a fuzzy inference combined with the traditional Kalman filter, obtaining the equivalent model and estimating in a fuzzy sense the gains matrix A and the proper gain K into the traditional filter identification. The second develops a direct estimation in state space, describing an EANN using the expected value and the recursive description of the gains estimation. Finally, a comparison of both descriptions is performed; highlighting the analytical method describes the neural net coefficients in a direct form, whereas the other technique requires selecting into the Knowledge Base (KB the factors based on the functional error and the reference signal built with the past information of the system.
Precipitation Estimation from Remotely Sensed Data Using Deep Neural Network
Tao, Y.; Gao, X.; Hsu, K. L.; Sorooshian, S.; Ihler, A.
2015-12-01
This research develops a precipitation estimation system from remote sensed data using state-of-the-art machine learning algorithms. Compared to ground-based precipitation measurements, satellite-based precipitation estimation products have advantages of temporal resolution and spatial coverage. Also, the massive amount of satellite data contains various measures related to precipitation formation and development. On the other hand, deep learning algorithms were newly developed in the area of machine learning, which was a breakthrough to deal with large and complex dataset, especially to image data. Here, we attempt to engage deep learning techniques to provide hourly precipitation estimation from satellite information, such as long wave infrared data. The brightness temperature data from infrared data is considered to contain cloud information. Radar stage IV dataset is used as ground measurement for parameter calibration. Stacked denoising auto-encoders (SDAE) is applied here to build a 4-layer neural network with 1000 hidden nodes for each hidden layer. SDAE involves two major steps: (1) greedily pre-training each layer as a denoising auto-encoder using the outputs of previous trained hidden layer output starting from visible layer to initialize parameters; (2) fine-tuning the whole deep neural network with supervised criteria. The results are compared with satellite precipitation product PERSIANN-CCS (Precipitation Estimation from Remotely Sensed Imagery using an Artificial Neural Network Cloud Classification System). Based on the results, we have several valuable conclusions: By properly training the neural network, it is able to extract useful information for precipitation estimation. For example, it can reduce the mean squared error of the precipitation by 58% for the summer season in the central United States of the validation period. The SDAE method captures the shape of the precipitation from the cloud shape better compared to the CCS product. Design of
Estimation of Effectivty Connectivity via Data-Driven Neural Modeling
Directory of Open Access Journals (Sweden)
Dean Robert Freestone
2014-11-01
Full Text Available This research introduces a new method for functional brain imaging via a process of model inversion. By estimating parameters of a computational model, we are able to track effective connectivity and mean membrane potential dynamics that cannot be directly measured using electrophysiological measurements alone. The ability to track the hidden aspects of neurophysiology will have a profound impact on the way we understand and treat epilepsy. For example, under the assumption the model captures the key features of the cortical circuits of interest, the framework will provide insights into seizure initiation and termination on a patient-specific basis. It will enable investigation into the effect a particular drug has on specific neural populations and connectivity structures using minimally invasive measurements. The method is based on approximating brain networks using an interconnected neural population model. The neural population model is based on a neural mass model that describes the functional activity of the brain, capturing the mesoscopic biophysics and anatomical structure. The model is made subject-specific by estimating the strength of intra-cortical connections within a region and inter-cortical connections between regions using a novel Kalman filtering method. We demonstrate through simulation how the framework can be used the track the mechanisms involved in seizure initiation and termination.
Artificial neural networks for stiffness estimation in magnetic resonance elastography.
Murphy, Matthew C; Manduca, Armando; Trzasko, Joshua D; Glaser, Kevin J; Huston, John; Ehman, Richard L
2017-11-28
To investigate the feasibility of using artificial neural networks to estimate stiffness from MR elastography (MRE) data. Artificial neural networks were fit using model-based training patterns to estimate stiffness from images of displacement using a patch size of ∼1 cm in each dimension. These neural network inversions (NNIs) were then evaluated in a set of simulation experiments designed to investigate the effects of wave interference and noise on NNI accuracy. NNI was also tested in vivo, comparing NNI results against currently used methods. In 4 simulation experiments, NNI performed as well or better than direct inversion (DI) for predicting the known stiffness of the data. Summary NNI results were also shown to be significantly correlated with DI results in the liver (R 2 = 0.974) and in the brain (R 2 = 0.915), and also correlated with established biological effects including fibrosis stage in the liver and age in the brain. Finally, repeatability error was lower in the brain using NNI compared to DI, and voxel-wise modeling using NNI stiffness maps detected larger effects than using DI maps with similar levels of smoothing. Artificial neural networks represent a new approach to inversion of MRE data. Summary results from NNI and DI are highly correlated and both are capable of detecting biologically relevant signals. Preliminary evidence suggests that NNI stiffness estimates may be more resistant to noise than an algebraic DI approach. Taken together, these results merit future investigation into NNIs to improve the estimation of stiffness in small regions. Magn Reson Med, 2017. © 2017 International Society for Magnetic Resonance in Medicine. © 2017 International Society for Magnetic Resonance in Medicine.
An artificial neural network for estimating haplotype frequencies
Baechle Daniel; Cartier Kevin C
2005-01-01
Abstract The problem of estimating haplotype frequencies from population data has been considered by numerous investigators, resulting in a wide variety of possible algorithmic and statistical solutions. We propose a relatively unique approach that employs an artificial neural network (ANN) to predict the most likely haplotype frequencies from a sample of population genotype data. Through an innovative ANN design for mapping genotype patterns to diplotypes, we have produced a prototype that d...
Convolutional neural networks for estimating spatially distributed evapotranspiration
García-Pedrero, Angel M.; Gonzalo-Martín, Consuelo; Lillo-Saavedra, Mario F.; Rodriguéz-Esparragón, Dionisio; Menasalvas, Ernestina
2017-10-01
Efficient water management in agriculture requires an accurate estimation of evapotranspiration (ET). There are available several balance energy surface models that provide a daily ET estimation (ETd) spatially and temporarily distributed for different crops over wide areas. These models need infrared thermal spectral band (gathered from remotely sensors) to estimate sensible heat flux from the surface temperature. However, this spectral band is not available for most current operational remote sensors. Even though the good results provided by machine learning (ML) methods in many different areas, few works have applied these approaches for forecasting distributed ETd on space and time when aforementioned information is missing. However, these methods do not exploit the land surface characteristics and the relationships among land covers producing estimation errors. In this work, we have developed and evaluated a methodology that provides spatial distributed estimates of ETd without thermal information by means of Convolutional Neural Networks.
An artificial neural network for estimating haplotype frequencies.
Cartier, Kevin C; Baechle, Daniel
2005-12-30
The problem of estimating haplotype frequencies from population data has been considered by numerous investigators, resulting in a wide variety of possible algorithmic and statistical solutions. We propose a relatively unique approach that employs an artificial neural network (ANN) to predict the most likely haplotype frequencies from a sample of population genotype data. Through an innovative ANN design for mapping genotype patterns to diplotypes, we have produced a prototype that demonstrates the feasibility of this approach, with provisional results that correlate well with estimates produced by the expectation maximization algorithm for haplotype frequency estimation. Given the computational demands of estimating haplotype frequencies for 20 or more single-nucleotide polymorphisms, the ANN approach is promising because its design fits well with parallel computing architectures.
Estimating Type Ia Supernova Metallicities Using Neural Networks
Villar, V. Ashley
2017-01-01
Normal Type Ia supernovae (SNe) can be used as standardizable candles because their progenitors, white dwarfs, are a fairly homogenous class of objects. However, intrinsic variability in these events arise from a number of factors, including metallicity. Recent studies have investigated the effects of metallicity on Type Ia SNe observables from both a theoretical approach, by tuning model metallicity to analyze spectral features, and an observational approach, by studying the effect of host metallicity on light curves. In this work, we take a new, data-driven approach to the problem. Inspired by the success of neural networks in the field of image processing, we aim to estimate the metallicities of Type Ia SNe progenitors from their near-maximum spectra using feed-forward neural networks. We first collect a sample of near-maximum Type Ia SNe spectra from the literature to be smoothed and down-sampled. We then estimate the metallicities of the SNe hosts using the B-band magnitudes. We build a multilayer perceptron to generate a model that takes as input the down-sampled spectra and returns a scalar metallicity. Finally, we discuss basic considerations to be taken when working with spectral (as opposed to image) data using neural networks.
Directory of Open Access Journals (Sweden)
Ramin Jaberi
2017-12-01
Full Text Available Purpose : Intra-fractional organs at risk (OARs deformations can lead to dose variation during image-guided adaptive brachytherapy (IGABT. The aim of this study was to modify the final accepted brachytherapy treatment plan to dosimetrically compensate for these intra-fractional organs-applicators position variations and, at the same time, fulfilling the dosimetric criteria. Material and methods : Thirty patients with locally advanced cervical cancer, after external beam radiotherapy (EBRT of 45-50 Gy over five to six weeks with concomitant weekly chemotherapy, and qualified for intracavitary high-dose-rate (HDR brachytherapy with tandem-ovoid applicators were selected for this study. Second computed tomography scan was done for each patient after finishing brachytherapy treatment with applicators in situ. Artificial neural networks (ANNs based models were used to predict intra-fractional OARs dose-volume histogram parameters variations and propose a new final plan. Results : A model was developed to estimate the intra-fractional organs dose variations during gynaecological intracavitary brachytherapy. Also, ANNs were used to modify the final brachytherapy treatment plan to compensate dosimetrically for changes in ‘organs-applicators’, while maintaining target dose at the original level. Conclusions : There are semi-automatic and fast responding models that can be used in the routine clinical workflow to reduce individually IGABT uncertainties. These models can be more validated by more patients’ plans to be able to serve as a clinical tool.
Estimation of Handling Flexible Cystoscope Using Neural Network
Nakamura, Munehiro; Kanaya, Jiro; Kimura, Haruhiko
This paper presents a system for estimating handling of a flexible cystoscope in bladder checkup. In the checkup, all the seven parts of the bladder wall can be observed with a flexible cystoscope. However, since the shape and color of the parts are depended on participants and five of the parts have same shape and color, it is difficult to grasp the location that the cystoscope is projecting. For the reason, it is often happened that urologists can not confirm that all the parts were observed. Since urologists conduct the checkup by handling the cystoscope and watching movement of the location that the cystoscope is projecting, reproducing the checkup by a computer would recognize oversights. And then, the proposed method extracts movement of the location and estimates the handling by the neural network. The experimental results of the estimations for 9 videos recorded in the checkup have shown a possibility of reproducing the checkup.
Gupta, S. M.; Malmgren, B. A.
2015-12-01
The western Pacific water enters into the Timor Sea (tropical Indian Ocean) by the thermohaline conveyor belt, and this region is under the influence of the SW monsoon. The higher precipitation during the monsoon rains lower the surface salinity in the north-eastern Indian Ocean towards the Bay of Bengal; whereas, the Arabian Sea remains highly saline due to higher evaporation in the region surrounding Arabian deserts. The salinity contrast in the northern Indian Ocean is very unique, and the radiolarian micro-zooplanktons living in the surface water serve a very good proxy for the monsoonal changes in the surface sea-water temperature (SST) and salinity in the geological past. We studied radiolarian faunal variation in the core MD01-2378, located at ~13oS and ~121oE (1783 m water depth), at the inlet of the thermohaline circulation into the Timor Sea. We applied the modern radiolarian based artificial neural networks (ANNs) (Gupta and Malmgren, 2009) to derive the SST and salinity during August-October for the last 140 ka (the full last glacial cycle). Based on the mean estimates of the 10 ANNs, the root mean square error in prediction (RMSEP) for SST is ~1.4oC with correlation between observed and estimated values r=0.98 (Gupta and Malmgren, 2009). Similarly, the RMSEP is 0.3 psu (r=0.94) for the salinity estimates. We derived paleo-SSTs and salinity values using modern radiolarian ANNs and the fossil radiolarian data generated from the core for the last 140-ka (Fig.1). The age model of the core is based on δ18O benthic oxygen isotope stratigraphy and 21 AMS 14C ages up to ~30-ka (Holbourn et al., 2005). Paleo SST-summer varied between 22-28.5oC, and it is in very good agreement with the δ18O benthic record of Holbourn et al. (2005) defining the Last Glacial Maximum (~24 ka) and the Eemian (~125 ka) stages. The salinity fluctuated between 34-35 psu, and compared well with oxygen isotope record representing the LGM and Eemian periods. We gratefully acknowledge
Ibarra, Juan G.; Tao, Yang; Xin, Hongwei
2000-11-01
A noninvasive method for the estimation of internal temperature in chicken meat immediately following cooking is proposed. The external temperature from IR images was correlated with measured internal temperature through a multilayer neural network. To provide inputs for the network, time series experiments were conducted to obtain simultaneous observations of internal and external temperatures immediately after cooking during the cooling process. An IR camera working at the spectral band of 3.4 to 5.0 micrometers registered external temperature distributions without the interference of close-to-oven environment, while conventional thermocouples registered internal temperatures. For an internal temperature at a given time, simultaneous and lagged external temperature observations were used as the input of the neural network. Based on practical and statistical considerations, a criterion is established to reduce the nodes in the neural network input. The combined method was able to estimate internal temperature for times between 0 and 540 s within a standard error of +/- 1.01 degree(s)C, and within an error of +/- 1.07 degree(s)C for short times after cooking (3 min), with two thermograms at times t and t+30s. The method has great potential for monitoring of doneness of chicken meat in conveyor belt type cooking and can be used as a platform for similar studies in other food products.
Hybrid fault diagnosis of nonlinear systems using neural parameter estimators.
Sobhani-Tehrani, E; Talebi, H A; Khorasani, K
2014-02-01
This paper presents a novel integrated hybrid approach for fault diagnosis (FD) of nonlinear systems taking advantage of both the system's mathematical model and the adaptive nonlinear approximation capability of computational intelligence techniques. Unlike most FD techniques, the proposed solution simultaneously accomplishes fault detection, isolation, and identification (FDII) within a unified diagnostic module. At the core of this solution is a bank of adaptive neural parameter estimators (NPEs) associated with a set of single-parameter fault models. The NPEs continuously estimate unknown fault parameters (FPs) that are indicators of faults in the system. Two NPE structures, series-parallel and parallel, are developed with their exclusive set of desirable attributes. The parallel scheme is extremely robust to measurement noise and possesses a simpler, yet more solid, fault isolation logic. In contrast, the series-parallel scheme displays short FD delays and is robust to closed-loop system transients due to changes in control commands. Finally, a fault tolerant observer (FTO) is designed to extend the capability of the two NPEs that originally assumes full state measurements for systems that have only partial state measurements. The proposed FTO is a neural state estimator that can estimate unmeasured states even in the presence of faults. The estimated and the measured states then comprise the inputs to the two proposed FDII schemes. Simulation results for FDII of reaction wheels of a three-axis stabilized satellite in the presence of disturbances and noise demonstrate the effectiveness of the proposed FDII solutions under partial state measurements. Copyright © 2013 Elsevier Ltd. All rights reserved.
Wang, Yunzhi; Qiu, Yuchen; Thai, Theresa; Moore, Kathleen; Liu, Hong; Zheng, Bin
2017-06-01
Accurately assessment of adipose tissue volume inside a human body plays an important role in predicting disease or cancer risk, diagnosis and prognosis. In order to overcome limitation of using only one subjectively selected CT image slice to estimate size of fat areas, this study aims to develop and test a computer-aided detection (CAD) scheme based on deep learning technique to automatically segment subcutaneous fat areas (SFA) and visceral fat areas (VFA) depicting on volumetric CT images. A retrospectively collected CT image dataset was divided into two independent training and testing groups. The proposed CAD framework consisted of two steps with two convolution neural networks (CNNs) namely, Selection-CNN and Segmentation-CNN. The first CNN was trained using 2,240 CT slices to select abdominal CT slices depicting SFA and VFA. The second CNN was trained with 84,000pixel patches and applied to the selected CT slices to identify fat-related pixels and assign them into SFA and VFA classes. Comparing to the manual CT slice selection and fat pixel segmentation results, the accuracy of CT slice selection using the Selection-CNN yielded 95.8%, while the accuracy of fat pixel segmentation using the Segmentation-CNN was 96.8%. This study demonstrated the feasibility of applying a new deep learning based CAD scheme to automatically recognize abdominal section of human body from CT scans and segment SFA and VFA from volumetric CT data with high accuracy or agreement with the manual segmentation results. Copyright © 2017 Elsevier B.V. All rights reserved.
Directory of Open Access Journals (Sweden)
Bahita Mohamed
2011-01-01
Full Text Available In this work, we introduce an adaptive neural network controller for a class of nonlinear systems. The approach uses two Radial Basis Functions, RBF networks. The first RBF network is used to approximate the ideal control law which cannot be implemented since the dynamics of the system are unknown. The second RBF network is used for on-line estimating the control gain which is a nonlinear and unknown function of the states. The updating laws for the combined estimator and controller are derived through Lyapunov analysis. Asymptotic stability is established with the tracking errors converging to a neighborhood of the origin. Finally, the proposed method is applied to control and stabilize the inverted pendulum system.
Estimation of Minimum DNBR Using Cascaded Fuzzy Neural Networks
Energy Technology Data Exchange (ETDEWEB)
Kim, Dong Yeong; Yoo, Kwae Hwan; Na, Man Gyun [Chosun University, Gwangju (Korea, Republic of)
2015-05-15
This phenomenon of boiling crisis is called a departure from nucleate boiling (DNB). The DNB phenomena can influence the fuel cladding and fuel pellets. The DNB ratio (DNBR) is defined as the ratio of the expected DNB heat flux to the actual fuel rod heat flux. Since it is very important to monitor and predict the minimum DNBR in a reactor core to prevent the boiling crisis and clad melting, a number of researches have been conducted to predict DNBR values. The aim of this study is to estimate the minimum DNBR in a reactor core using the measured signals of the reactor coolant system (RCS) by applying cascaded fuzzy neural networks (CFNN) according to operating conditions. Reactor core monitoring and protection systems require minimum DNBR prediction. The CFNN can be used to optimize the minimum DNBR value through the process of adding fuzzy neural networks (FNN) repeatedly. The proposed algorithm is trained by using the data set prepared for training (development data) and verified by using another data set different (independent) from the development data. The developed CFNN models were applied to the first fuel cycle of OPR1000. The RMS errors are 0.23% and 0.12% for the positive and negative ASI, respectively.
Energy Technology Data Exchange (ETDEWEB)
Loddenkoetter, Thomas
2012-08-15
In order to provide discrimination between the Wt-channel signal and its backgrounds for analyses that try to measure single top-quark production in the Wt-channel, a kinematic fit to the lepton+jets decay mode of the Wt-channel has been implemented using the KLFitter package. The fit has been validated by studying its performance in terms of the efficiency of the fit to correctly assign the final-state quarks of the fit model to the measured jets as a function of various parameters, as well as the improvement of the energy resolutions of the fitted particles due to the fit. By combining the output variables of the kinematic fitter using neural networks, it has been shown that the fit results are suitable to identify the decay mode of the top quark in Wt events and to identify whether the kinematic fit succeeded in correctly assigning the final-state quarks to the measured jets. In order to demonstrate the value of the kinematic fit for analysis, another neural network - again using strictly results of the kinematic fit as input - has been trained to separate to the Wt-channel signal from its backgrounds. A separation power comparable to a conventional neural-network-based Wt-channel analysis has been achieved.
Estimation of LOCA break size using cascaded Fuzzy neural networks
Energy Technology Data Exchange (ETDEWEB)
Choi, Geon Pil; Yoo, Kwae Hwan; Back, Ju Hyun; Na, Man Gyun [Dept. of Nuclear Engineering, Chosun University, Gwangju (Korea, Republic of)
2017-04-15
Operators of nuclear power plants may not be equipped with sufficient information during a loss-of-coolant accident (LOCA), which can be fatal, or they may not have sufficient time to analyze the information they do have, even if this information is adequate. It is not easy to predict the progression of LOCAs in nuclear power plants. Therefore, accurate information on the LOCA break position and size should be provided to efficiently manage the accident. In this paper, the LOCA break size is predicted using a cascaded fuzzy neural network (CFNN) model. The input data of the CFNN model are the time-integrated values of each measurement signal for an initial short-time interval after a reactor scram. The training of the CFNN model is accomplished by a hybrid method combined with a genetic algorithm and a least squares method. As a result, LOCA break size is estimated exactly by the proposed CFNN model.
Estimating plant root water uptake using a neural network approach
DEFF Research Database (Denmark)
Qiao, D M; Shi, H B; Pang, H B
2010-01-01
Water uptake by plant roots is an important process in the hydrological cycle, not only for plant growth but also for the role it plays in shaping microbial community and bringing in physical and biochemical changes to soils. The ability of roots to extract water is determined by combined soil...... and plant characteristics, and how to model it has been of interest for many years. Most macroscopic models for water uptake operate at soil profile scale under the assumption that the uptake rate depends on root density and soil moisture. Whilst proved appropriate, these models need spatio-temporal root...... but has not yet been addressed. This paper presents and tests such an approach. The method is based on a neural network model, estimating the water uptake using different types of data that are easy to measure in the field. Sunflower grown in a sandy loam subjected to water stress and salinity was taken...
Estimation of LOCA Break Size Using Cascaded Fuzzy Neural Networks
Directory of Open Access Journals (Sweden)
Geon Pil Choi
2017-04-01
Full Text Available Operators of nuclear power plants may not be equipped with sufficient information during a loss-of-coolant accident (LOCA, which can be fatal, or they may not have sufficient time to analyze the information they do have, even if this information is adequate. It is not easy to predict the progression of LOCAs in nuclear power plants. Therefore, accurate information on the LOCA break position and size should be provided to efficiently manage the accident. In this paper, the LOCA break size is predicted using a cascaded fuzzy neural network (CFNN model. The input data of the CFNN model are the time-integrated values of each measurement signal for an initial short-time interval after a reactor scram. The training of the CFNN model is accomplished by a hybrid method combined with a genetic algorithm and a least squares method. As a result, LOCA break size is estimated exactly by the proposed CFNN model.
Deep convolutional neural network approach for forehead tissue thickness estimation
Directory of Open Access Journals (Sweden)
Manit Jirapong
2017-09-01
Full Text Available In this paper, we presented a deep convolutional neural network (CNN approach for forehead tissue thickness estimation. We use down sampled NIR laser backscattering images acquired from a novel marker-less near-infrared laser-based head tracking system, combined with the beam’s incident angle parameter. These two-channel augmented images were constructed for the CNN input, while a single node output layer represents the estimated value of the forehead tissue thickness. The models were – separately for each subject – trained and tested on datasets acquired from 30 subjects (high resolution MRI data is used as ground truth. To speed up training, we used a pre-trained network from the first subject to bootstrap training for each of the other subjects. We could show a clear improvement for the tissue thickness estimation (mean RMSE of 0.096 mm. This proposed CNN model outperformed previous support vector regression (mean RMSE of 0.155 mm or Gaussian processes learning approaches (mean RMSE of 0.114 mm and eliminated their restrictions for future research.
Zhang, B; Liang, X L; Gao, H Y; Ye, L S; Wang, Y G
2016-05-13
We evaluated the application of three machine learning algorithms, including logistic regression, support vector machine and back-propagation neural network, for diagnosing congenital heart disease and colorectal cancer. By inspecting related serum tumor marker levels in colorectal cancer patients and healthy subjects, early diagnosis models for colorectal cancer were built using three machine learning algorithms to assess their corresponding diagnostic values. Except for serum alpha-fetoprotein, the levels of 11 other serum markers of patients in the colorectal cancer group were higher than those in the benign colorectal cancer group (P model and back-propagation, a neural network diagnosis model was built with diagnostic accuracies of 82 and 75%, sensitivities of 85 and 80%, and specificities of 80 and 70%, respectively. Colorectal cancer diagnosis models based on the three machine learning algorithms showed high diagnostic value and can help obtain evidence for the early diagnosis of colorectal cancer.
Improved estimation of energy expenditure by artificial neural network modeling.
Hay, Dean Charles; Wakayama, Akinobu; Sakamura, Ken; Fukashiro, Senshi
2008-12-01
Estimation of energy expenditure in daily living conditions can be a tool for clinical assessment of health status, as well as a self-measure of lifestyle and general activity levels. Criterion measures are either prohibitively expensive or restricted to laboratory settings. Portable devices (heart rate monitors, pedometers) have gained recent popularity, but accuracy of the prediction equations remains questionable. This study applied an artificial neural network modeling approach to the problem of estimating energy expenditure with different dynamic inputs (accelerometry, heart rate above resting (HRar), and electromyography (EMG)). Nine feed-forward back-propagation models were trained, with the goal of minimizing the mean squared error (MSE) of the training datasets. Model 1 (accelerometry only) and model 2 (HRar only) performed poorly and had significantly greater MSE than all other models (p energy expenditure for models 3 to 9 ranged from 0.745 to 0.817. Analysis of mean error within specific movement categories indicates that EMG models may be better at predicting higher-intensity energy expenditure, but combined accelerometry and HRar provides an economical solution, with sufficient accuracy.
Total alkalinity estimation using MLR and neural network techniques
Velo, A.; Pérez, F. F.; Tanhua, T.; Gilcoto, M.; Ríos, A. F.; Key, R. M.
2013-02-01
During the last decade, two important collections of carbon relevant hydrochemical data have become available: GLODAP and CARINA. These collections comprise a synthesis of bottle data for all ocean depths from many cruises collected over several decades. For a majority of the cruises at least two carbon parameters were measured. However, for a large number of stations, samples or even cruises, the carbonate system is under-determined (i.e., only one or no carbonate parameter was measured) resulting in data gaps for the carbonate system in these collections. A method for filling these gaps would be very useful, as it would help with estimations of the anthropogenic carbon (Cant) content or quantification of oceanic acidification. The aim of this work is to apply and describe, a 3D moving window multilinear regression algorithm (MLR) to fill gaps in total alkalinity (AT) of the CARINA and GLODAP data collections for the Atlantic. In addition to filling data gaps, the estimated AT values derived from the MLR are useful in quality control of the measurements of the carbonate system, as they can aid in the identification of outliers. For comparison, a neural network algorithm able to perform non-linear predictions was also designed. The goal here was to design an alternative approach to accomplish the same task of filling AT gaps. Both methods return internally consistent results, thereby giving confidence in our approach.
Koller, Michal
Remote sensing is one of the major data acquisition tools available to rapidly acquire soil and plant related information over a wide area for use in precision agriculture. Green canopy has very specific reflectance characteristics distinguishing it from other materials such as soil and dry vegetative matter. Reflectance values in red (R) and near infra-red (NIR) spectral bands have been widely used for calculating normalized difference vegetation index (NDVI). Many researchers have related NDVI values to plant vigor, water stress, leaf area index (LAI) and/or yield. However, vegetative indices such as NDVI are usually sensitive to background reflectance characteristics. Often soil adjusted vegetation indices (SAVI) are used to minimize the background effect. In this study we have developed a relationship between the processing tomato yield and SAVI based on the R and NIR reflectance. Eight three band (R, NIR and green) aerial images were obtained at approximately two-week intervals during the 2000 processing tomato growing season. These images were analyzed to obtain SAVI values which were in turn related to LAI using regression techniques. A tuned neural network was developed to predict daily LAI values based on the biweekly experimental LAI values derived from aerial images. The coefficients of multiple determination between the actual LAI and neural network predicted LAI values were greater than 0.96 for all 56 grid points. The LAI values were numerically integrated over the whole growing season to obtain cumulative leaf area index days (CLAID). The CLAID values predicted from the neural network correlated very well with experimentally derived CLAID values (coefficient of determination, r2 = 0.83) indicating that the neural network model simulated processing tomato growth well. A crop growth model that was capable of predicting crop yield based on neural network predicted LAI values and CIMIS weather data was developed. Although predicted yield tended to be low
National Research Council Canada - National Science Library
Friman, Henrik
2006-01-01
... (extended from Leavitt, 1965). This text identifies aspects of network-based effectiveness that can benefit from a better understanding of leadership and management development of people, procedures, technology, and organizations...
Directory of Open Access Journals (Sweden)
Mohammad Heidari
2014-01-01
Full Text Available The static pull-in instability of beam-type microelectromechanical systems (MEMS is theoretically investigated. Two engineering cases including cantilever and double cantilever microbeam are considered. Considering the midplane stretching as the source of the nonlinearity in the beam behavior, a nonlinear size-dependent Euler-Bernoulli beam model is used based on a modified couple stress theory, capable of capturing the size effect. By selecting a range of geometric parameters such as beam lengths, width, thickness, gaps, and size effect, we identify the static pull-in instability voltage. A MAPLE package is employed to solve the nonlinear differential governing equations to obtain the static pull-in instability voltage of microbeams. Radial basis function artificial neural network with two functions has been used for modeling the static pull-in instability of microcantilever beam. The network has four inputs of length, width, gap, and the ratio of height to scale parameter of beam as the independent process variables, and the output is static pull-in voltage of microbeam. Numerical data, employed for training the network, and capabilities of the model have been verified in predicting the pull-in instability behavior. The output obtained from neural network model is compared with numerical results, and the amount of relative error has been calculated. Based on this verification error, it is shown that the radial basis function of neural network has the average error of 4.55% in predicting pull-in voltage of cantilever microbeam. Further analysis of pull-in instability of beam under different input conditions has been investigated and comparison results of modeling with numerical considerations shows a good agreement, which also proves the feasibility and effectiveness of the adopted approach. The results reveal significant influences of size effect and geometric parameters on the static pull-in instability voltage of MEMS.
Heidari, Mohammad; Heidari, Ali; Homaei, Hadi
2014-01-01
The static pull-in instability of beam-type microelectromechanical systems (MEMS) is theoretically investigated. Two engineering cases including cantilever and double cantilever microbeam are considered. Considering the midplane stretching as the source of the nonlinearity in the beam behavior, a nonlinear size-dependent Euler-Bernoulli beam model is used based on a modified couple stress theory, capable of capturing the size effect. By selecting a range of geometric parameters such as beam lengths, width, thickness, gaps, and size effect, we identify the static pull-in instability voltage. A MAPLE package is employed to solve the nonlinear differential governing equations to obtain the static pull-in instability voltage of microbeams. Radial basis function artificial neural network with two functions has been used for modeling the static pull-in instability of microcantilever beam. The network has four inputs of length, width, gap, and the ratio of height to scale parameter of beam as the independent process variables, and the output is static pull-in voltage of microbeam. Numerical data, employed for training the network, and capabilities of the model have been verified in predicting the pull-in instability behavior. The output obtained from neural network model is compared with numerical results, and the amount of relative error has been calculated. Based on this verification error, it is shown that the radial basis function of neural network has the average error of 4.55% in predicting pull-in voltage of cantilever microbeam. Further analysis of pull-in instability of beam under different input conditions has been investigated and comparison results of modeling with numerical considerations shows a good agreement, which also proves the feasibility and effectiveness of the adopted approach. The results reveal significant influences of size effect and geometric parameters on the static pull-in instability voltage of MEMS.
Energy Technology Data Exchange (ETDEWEB)
Doh, Jaeh Yeok; Lee, Jong Soo [Yonsei University, Seoul (Korea, Republic of); Lee, Seung Uk [Gyeongbuk Hybrid Technology Institute, Yeongcheon (Korea, Republic of)
2016-03-15
In this study, a Back-propagation neural network (BPN) is employed to conduct an approximation of a true stress-strain curve using the load-displacement experimental data of DP590, a high-strength material used in automobile bodies and chassis. The optimized interconnection weights are obtained with hidden layers and output layers of the BPN through intelligent learning and training of the experimental data; by using these weights, a mathematical model of the material's behavior is suggested through this feed-forward neural network. Generally, the material properties from the tensile test cannot be acquired until the fracture regions, since it is difficult to measure the cross-section area of a specimen after diffusion necking. For this reason, the plastic properties of the true stress-strain are extrapolated using the weighted-average method after diffusion necking. The accuracies of BPN-based meta-models for predicting material properties are validated in terms of the Root mean square error (RMSE). By applying the approximate material properties, the reliable finite element solution can be obtained to realize the different shapes of the finite element models. Furthermore, the sensitivity analysis of the approximate meta-model is performed using the first-order approximate derivatives of the BPN and is compared with the results of the finite difference method. In addition, we predict the tension velocity's effect on the material property through a first-order sensitivity analysis.
Directory of Open Access Journals (Sweden)
Shazwani Samson
Full Text Available A predictive model of a virgin coconut oil (VCO nanoemulsion system for the topical delivery of copper peptide (an anti-aging compound was developed using an artificial neural network (ANN to investigate the factors that influence particle size. Four independent variables including the amount of VCO, Tween 80: Pluronic F68 (T80:PF68, xanthan gum and water were the inputs whereas particle size was taken as the response for the trained network. Genetic algorithms (GA were used to model the data which were divided into training sets, testing sets and validation sets. The model obtained indicated the high quality performance of the neural network and its capability to identify the critical composition factors for the VCO nanoemulsion. The main factor controlling the particle size was found out to be xanthan gum (28.56% followed by T80:PF68 (26.9%, VCO (22.8% and water (21.74%. The formulation containing copper peptide was then successfully prepared using optimum conditions and particle sizes of 120.7 nm were obtained. The final formulation exhibited a zeta potential lower than -25 mV and showed good physical stability towards centrifugation test, freeze-thaw cycle test and storage at temperature 25°C and 45°C.
Hansen, T. M.; Cordua, K. S.
2017-12-01
Probabilistically formulated inverse problems can be solved using Monte Carlo-based sampling methods. In principle, both advanced prior information, based on for example, complex geostatistical models and non-linear forward models can be considered using such methods. However, Monte Carlo methods may be associated with huge computational costs that, in practice, limit their application. This is not least due to the computational requirements related to solving the forward problem, where the physical forward response of some earth model has to be evaluated. Here, it is suggested to replace a numerical complex evaluation of the forward problem, with a trained neural network that can be evaluated very fast. This will introduce a modeling error that is quantified probabilistically such that it can be accounted for during inversion. This allows a very fast and efficient Monte Carlo sampling of the solution to an inverse problem. We demonstrate the methodology for first arrival traveltime inversion of crosshole ground penetrating radar data. An accurate forward model, based on 2-D full-waveform modeling followed by automatic traveltime picking, is replaced by a fast neural network. This provides a sampling algorithm three orders of magnitude faster than using the accurate and computationally expensive forward model, and also considerably faster and more accurate (i.e. with better resolution), than commonly used approximate forward models. The methodology has the potential to dramatically change the complexity of non-linear and non-Gaussian inverse problems that have to be solved using Monte Carlo sampling techniques.
Directory of Open Access Journals (Sweden)
Dongxiao Niu
2017-12-01
Full Text Available Accurate and stable prediction of icing thickness on transmission lines is of great significance for ensuring the safe operation of the power grid. In order to improve the accuracy and stability of icing prediction, an innovative prediction model based on the generalized regression neural network (GRNN and the fruit fly optimization algorithm (FOA is proposed. Firstly, a feature selection method based on the data inconsistency rate (IR is adopted to select the optimal feature, which aims to reduce redundant input vectors. Then, the fruit FOA is utilized for optimization of smoothing factor for the GRNN. Lastly, the icing forecasting method FOA-IR-GRNN is established. Two cases in different locations and different months are selected to validate the proposed model. The results indicate that the new hybrid FOA-IR-GRNN model presents better accuracy, robustness, and generality in icing forecasting.
Prasad, Archana; Prakash, Om; Mehrotra, Shakti; Khan, Feroz; Mathur, Ajay Kumar; Mathur, Archana
2017-01-01
An artificial neural network (ANN)-based modelling approach is used to determine the synergistic effect of five major components of growth medium (Mg, Cu, Zn, nitrate and sucrose) on improved in vitro biomass yield in multiple shoot cultures of Centella asiatica. The back propagation neural network (BPNN) was employed to predict optimal biomass accumulation in terms of growth index over a defined culture duration of 35 days. The four variable concentrations of five media components, i.e. MgSO 4 (0, 0.75, 1.5, 3.0 mM), ZnSO 4 (0, 15, 30, 60 μM), CuSO 4 (0, 0.05, 0.1, 0.2 μM), NO 3 (20, 30, 40, 60 mM) and sucrose (1, 3, 5, 7 %, w/v) were taken as inputs for the ANN model. The designed model was evaluated by performing three different sets of validation experiments that indicated a greater similarity between the target and predicted dataset. The results of the modelling experiment suggested that 1.5 mM Mg, 30 μM Zn, 0.1 μM Cu, 40 mM NO 3 and 6 % (w/v) sucrose were the respective optimal concentrations of the tested medium components for achieving maximum growth index of 1654.46 with high centelloside yield (62.37 mg DW/culture) in the cultured multiple shoots. This study can facilitate the generation of higher biomass of uniform, clean, good quality C. asiatica herb that can efficiently be utilized by pharmaceutical industries.
DEFF Research Database (Denmark)
Chon, K H; Hoyer, D; Armoundas, A A
1999-01-01
error is obtained by subtracting the corrupt signal of the estimated ARMA model obtained via the deterministic estimation step from the system output response. We present computer simulation examples to show the efficacy of the proposed stochastic recurrent neural network approach in obtaining accurate......In this study, we introduce a new approach for estimating linear and nonlinear stochastic autoregressive moving average (ARMA) model parameters, given a corrupt signal, using artificial recurrent neural networks. This new approach is a two-step approach in which the parameters of the deterministic...... part of the stochastic ARMA model are first estimated via a three-layer artificial neural network (deterministic estimation step) and then reestimated using the prediction error as one of the inputs to the artificial neural networks in an iterative algorithm (stochastic estimation step). The prediction...
Naushad, Shaik Mohammad; Ramaiah, M Janaki; Pavithrakumari, Manickam; Jayapriya, Jaganathan; Hussain, Tajamul; Alrokayan, Salman A; Gottumukkala, Suryanarayana Raju; Digumarti, Raghunadharao; Kutala, Vijay Kumar
2016-04-15
In the current study, an artificial neural network (ANN)-based breast cancer prediction model was developed from the data of folate and xenobiotic pathway genetic polymorphisms along with the nutritional and demographic variables to investigate how micronutrients modulate susceptibility to breast cancer. The developed ANN model explained 94.2% variability in breast cancer prediction. Fixed effect models of folate (400 μg/day) and B12 (6 μg/day) showed 33.3% and 11.3% risk reduction, respectively. Multifactor dimensionality reduction analysis showed the following interactions in responders to folate: RFC1 G80A × MTHFR C677T (primary), COMT H108L × CYP1A1 m2 (secondary), MTR A2756G (tertiary). The interactions among responders to B12 were RFC1G80A × cSHMT C1420T and CYP1A1 m2 × CYP1A1 m4. ANN simulations revealed that increased folate might restore ER and PR expression and reduce the promoter CpG island methylation of extra cellular superoxide dismutase and BRCA1. Dietary intake of folate appears to confer protection against breast cancer through its modulating effects on ER and PR expression and methylation of EC-SOD and BRCA1. Copyright © 2016 Elsevier B.V. All rights reserved.
Vairamuthu, G.; Thangagiri, B.; Sundarapandian, S.
2017-07-01
The present work investigates the effect of varying Nozzle Opening Pressures (NOP) from 220 bar to 250 bar on performance, emissions and combustion characteristics of Calophyllum inophyllum Methyl Ester (CIME) in a constant speed, Direct Injection (DI) diesel engine using Artificial Neural Network (ANN) approach. An ANN model has been developed to predict a correlation between specific fuel consumption (SFC), brake thermal efficiency (BTE), exhaust gas temperature (EGT), Unburnt hydrocarbon (UBHC), CO, CO2, NOx and smoke density using load, blend (B0 and B100) and NOP as input data. A standard Back-Propagation Algorithm (BPA) for the engine is used in this model. A Multi Layer Perceptron network (MLP) is used for nonlinear mapping between the input and the output parameters. An ANN model can predict the performance of diesel engine and the exhaust emissions with correlation coefficient (R2) in the range of 0.98-1. Mean Relative Errors (MRE) values are in the range of 0.46-5.8%, while the Mean Square Errors (MSE) are found to be very low. It is evident that the ANN models are reliable tools for the prediction of DI diesel engine performance and emissions. The test results show that the optimum NOP is 250 bar with B100.
Directory of Open Access Journals (Sweden)
Junya Lv
2017-01-01
Full Text Available The application of accurate constitutive relationship in finite element simulation would significantly contribute to accurate simulation results, which play critical roles in process design and optimization. In this investigation, the true stress-strain data of an Inconel 718 superalloy were obtained from a series of isothermal compression tests conducted in a wide temperature range of 1153–1353 K and strain rate range of 0.01–10 s−1 on a Gleeble 3500 testing machine (DSI, St. Paul, DE, USA. Then the constitutive relationship was modeled by an optimally-constructed and well-trained back-propagation artificial neural network (ANN. The evaluation of the ANN model revealed that it has admirable performance in characterizing and predicting the flow behaviors of Inconel 718 superalloy. Consequently, the developed ANN model was used to predict abundant stress-strain data beyond the limited experimental conditions and construct the continuous mapping relationship for temperature, strain rate, strain and stress. Finally, the constructed ANN was implanted in a finite element solver though the interface of “URPFLO” subroutine to simulate the isothermal compression tests. The results show that the integration of finite element method with ANN model can significantly promote the accuracy improvement of numerical simulations for hot forming processes.
Vairamuthu, G.; Thangagiri, B.; Sundarapandian, S.
2018-01-01
The present work investigates the effect of varying Nozzle Opening Pressures (NOP) from 220 bar to 250 bar on performance, emissions and combustion characteristics of Calophyllum inophyllum Methyl Ester (CIME) in a constant speed, Direct Injection (DI) diesel engine using Artificial Neural Network (ANN) approach. An ANN model has been developed to predict a correlation between specific fuel consumption (SFC), brake thermal efficiency (BTE), exhaust gas temperature (EGT), Unburnt hydrocarbon (UBHC), CO, CO2, NOx and smoke density using load, blend (B0 and B100) and NOP as input data. A standard Back-Propagation Algorithm (BPA) for the engine is used in this model. A Multi Layer Perceptron network (MLP) is used for nonlinear mapping between the input and the output parameters. An ANN model can predict the performance of diesel engine and the exhaust emissions with correlation coefficient (R2) in the range of 0.98-1. Mean Relative Errors (MRE) values are in the range of 0.46-5.8%, while the Mean Square Errors (MSE) are found to be very low. It is evident that the ANN models are reliable tools for the prediction of DI diesel engine performance and emissions. The test results show that the optimum NOP is 250 bar with B100.
Directory of Open Access Journals (Sweden)
Xiaofei Yan
2016-08-01
Full Text Available Diverse sensing techniques have been developed and combined with machine learning method for forest fire detection, but none of them referred to identifying smoldering and flaming combustion phases. This study attempts to real-time identify different combustion phases using a developed wireless sensor network (WSN-based multi-sensor system and artificial neural network (ANN. Sensors (CO, CO2, smoke, air temperature and relative humidity were integrated into one node of WSN. An experiment was conducted using burning materials from residual of forest to test responses of each node under no, smoldering-dominated and flaming-dominated combustion conditions. The results showed that the five sensors have reasonable responses to artificial forest fire. To reduce cost of the nodes, smoke, CO2 and temperature sensors were chiefly selected through correlation analysis. For achieving higher identification rate, an ANN model was built and trained with inputs of four sensor groups: smoke; smoke and CO2; smoke and temperature; smoke, CO2 and temperature. The model test results showed that multi-sensor input yielded higher predicting accuracy (≥82.5% than single-sensor input (50.9%–92.5%. Based on these, it is possible to reduce the cost with a relatively high fire identification rate and potential application of the system can be tested in future under real forest condition.
Yan, Xiaofei; Cheng, Hong; Zhao, Yandong; Yu, Wenhua; Huang, Huan; Zheng, Xiaoliang
2016-08-04
Diverse sensing techniques have been developed and combined with machine learning method for forest fire detection, but none of them referred to identifying smoldering and flaming combustion phases. This study attempts to real-time identify different combustion phases using a developed wireless sensor network (WSN)-based multi-sensor system and artificial neural network (ANN). Sensors (CO, CO₂, smoke, air temperature and relative humidity) were integrated into one node of WSN. An experiment was conducted using burning materials from residual of forest to test responses of each node under no, smoldering-dominated and flaming-dominated combustion conditions. The results showed that the five sensors have reasonable responses to artificial forest fire. To reduce cost of the nodes, smoke, CO₂ and temperature sensors were chiefly selected through correlation analysis. For achieving higher identification rate, an ANN model was built and trained with inputs of four sensor groups: smoke; smoke and CO₂; smoke and temperature; smoke, CO₂ and temperature. The model test results showed that multi-sensor input yielded higher predicting accuracy (≥82.5%) than single-sensor input (50.9%-92.5%). Based on these, it is possible to reduce the cost with a relatively high fire identification rate and potential application of the system can be tested in future under real forest condition.
Gonzalez Vilas, L.; Castro Fernandez, M.; Spyrakos, E.; Torres Palenzuela, J.
2013-08-01
In typical case 2 waters an accurate remote sensing retrieval of chlorophyll a (chla) is still challenging. There is a widespread understanding that universally applicable water constituent retrieval algorithms are currently not feasible, shifting the research focus to regionally specific implementations of powerful inversion methods. This study takes advantage of regionally specific chlorophyll a (chla) algorithms, which were developed by the authors of this abstract in previous works, and the characteristics of Medium Resolution Imaging Spectrometer (MERIS) in order to study harmful algal events in the optically complex waters of the Galician Rias (NW). Harmful algal events are a frequent phenomenon in this area with direct and indirect impacts to the mussel production that constitute a very important economic activity for the local community. More than 240 106 kg of mussel per year are produced in these highly primary productive upwelling systems. A MERIS archive from nine years (2003-2012) was analysed using regionally specific chla algorithms. The latter were developed based on Multilayer perceptron (MLP) artificial neural networks and fuzzy c-mean clustering techniques (FCM). FCM specifies zones (based on water leaving reflectances) where the retrieval algorithms normally provide more reliable results. Monthly chla anomalies and other statistics were calculated for the nine years MERIS archive. These results were then related to upwelling indices and other associated measurements to determine the driver forces for specific phytoplankton blooms. The distribution and changes of chla are also discussed.
A hybrid training method for neural energy estimation in calorimetry
Da Silva, P V M; Seixas, J
2001-01-01
A neural mapping is developed to improve the overall performance of Tilecal, which is the hadronic calorimeter of the ATLAS detector. Feeding the input nodes of a multilayer feedforward neural network with the energy values sampled by the calorimeter cells in beam tests, it is shown that the original energy scale of pion beams is reconstructed over a wide energy range and linearity is significantly improved. As it happens for classical methods, a compromise between nonlinearity correction and the optimization of the energy resolution of the detector has to be accomplished. A hybrid training method for the neural mapping is proposed to achieve this design goal. Using the backpropagation algorithm, the method intercalates an epoch of training steps, for which the neural mapping mainly focus on linearity correction, with another block of training steps, in which the original energy resolution obtained by linearly combining the calorimeter cells becomes the main target. (6 refs).
Wu, Liang; Wan, Congying; Wang, Supin; Wan, Mingxi
2013-03-01
To realize an accurate and automatic on/off control of electrolarynx (EL), an artificial neural network (ANN) was introduced for switch identification based on visual information of lips and implemented by an experimental system (ANN-EL). The objective was to confirm the feasibility of the ANN method and evaluate the performance of ANN-EL in Mandarin speech. Totally five volunteers (one laryngectomee and four normal speakers) participated in the whole process of experiments. First, trained ANN was tested to assess switch identification performance of ANN method. Then, voice initiation/termination time, speech fluency, and word intelligibility were measured and compared with button-EL and video-EL to evaluate on/off control performance of ANN-EL. The test showed that ANN method performed accurate switch identification (>99%). ANN-EL was as fast as normal voice and button-EL in onset control, but a little slower in offset control. ANN-EL could provide a fluent voice source with rare breaks (speech. The results also indicated that on/off control performance of ANN-EL had a significant impact on perception, lowering the word intelligibility compared with button-EL. However, the words produced by ANN-EL were more intelligible than video-EL by approximately 20%. The ANN method was proved feasible and effective for switch identification based on visual information of lips. The ANN-EL could provide an accurate on/off control for fluent Mandarin speech. Copyright © 2013 The Voice Foundation. Published by Mosby, Inc. All rights reserved.
Kang, Jin Kyu; Hong, Hyung Gil; Park, Kang Ryoung
2017-07-08
A number of studies have been conducted to enhance the pedestrian detection accuracy of intelligent surveillance systems. However, detecting pedestrians under outdoor conditions is a challenging problem due to the varying lighting, shadows, and occlusions. In recent times, a growing number of studies have been performed on visible light camera-based pedestrian detection systems using a convolutional neural network (CNN) in order to make the pedestrian detection process more resilient to such conditions. However, visible light cameras still cannot detect pedestrians during nighttime, and are easily affected by shadows and lighting. There are many studies on CNN-based pedestrian detection through the use of far-infrared (FIR) light cameras (i.e., thermal cameras) to address such difficulties. However, when the solar radiation increases and the background temperature reaches the same level as the body temperature, it remains difficult for the FIR light camera to detect pedestrians due to the insignificant difference between the pedestrian and non-pedestrian features within the images. Researchers have been trying to solve this issue by inputting both the visible light and the FIR camera images into the CNN as the input. This, however, takes a longer time to process, and makes the system structure more complex as the CNN needs to process both camera images. This research adaptively selects a more appropriate candidate between two pedestrian images from visible light and FIR cameras based on a fuzzy inference system (FIS), and the selected candidate is verified with a CNN. Three types of databases were tested, taking into account various environmental factors using visible light and FIR cameras. The results showed that the proposed method performs better than the previously reported methods.
Estimation of concrete compressive strength using artificial neural network
Directory of Open Access Journals (Sweden)
Kostić Srđan
2015-01-01
Full Text Available In present paper, concrete compressive strength is evaluated using back propagation feed-forward artificial neural network. Training of neural network is performed using Levenberg-Marquardt learning algorithm for four architectures of artificial neural networks, one, three, eight and twelve nodes in a hidden layer in order to avoid the occurrence of overfitting. Training, validation and testing of neural network is conducted for 75 concrete samples with distinct w/c ratio and amount of superplasticizer of melamine type. These specimens were exposed to different number of freeze/thaw cycles and their compressive strength was determined after 7, 20 and 32 days. The obtained results indicate that neural network with one hidden layer and twelve hidden nodes gives reasonable prediction accuracy in comparison to experimental results (R=0.965, MSE=0.005. These results of the performed analysis are further confirmed by calculating the standard statistical errors: the chosen architecture of neural network shows the smallest value of mean absolute percentage error (MAPE=, variance absolute relative error (VARE and median absolute error (MEDAE, and the highest value of variance accounted for (VAF.
Cui, Yaokui; Long, Di; Hong, Yang; Zeng, Chao; Zhou, Jie; Han, Zhongying; Liu, Ronghua; Wan, Wei
2016-12-01
Soil moisture is a key variable in the exchange of water and energy between the land surface and the atmosphere, especially over the Tibetan Plateau (TP) which is climatically and hydrologically sensitive as the Earth's 'third pole'. Large-scale spatially consistent and temporally continuous soil moisture datasets are of great importance to meteorological and hydrological applications, such as weather forecasting and drought monitoring. The Fengyun-3B Microwave Radiation Imager (FY-3B/MWRI) soil moisture product is a relatively new passive microwave product, with the satellite being launched on November 5, 2010. This study validates and reconstructs FY-3B/MWRI soil moisture across the TP. First, the validation is performed using in situ measurements within two in situ soil moisture measurement networks (1° × 1° and 0.25° × 0.25°), and also compared with the Essential Climate Variable (ECV) soil moisture product from multiple active and passive satellite soil moisture products using new merging procedures. Results show that the ascending FY-3B/MWRI product outperforms the descending product. The ascending FY-3B/MWRI product has almost the same correlation as the ECV product with the in situ measurements. The ascending FY-3B/MWRI product has better performance than the ECV product in the frozen season and under the lower NDVI condition. When the NDVI is higher in the unfrozen season, uncertainty in the ascending FY-3B/MWRI product increases with increasing NDVI, but it could still capture the variability in soil moisture. Second, the FY-3B/MWRI soil moisture product is subsequently reconstructed using the back-propagation neural network (BP-NN) based on reconstructed MODIS products, i.e., LST, NDVI, and albedo. The reconstruction method of generating the soil moisture product not only considers the relationship between the soil moisture and NDVI, LST, and albedo, but also the relationship between the soil moisture and four-dimensional variations using the
Wichaipanich, Noraset; Hozumi, Kornyanat; Supnithi, Pornchai; Tsugawa, Takuya
2017-06-01
This paper presents the development of Neural Network (NN) model for the prediction of the F2 layer critical frequency (foF2) at three ionosonde stations near the magnetic equator of Southeast Asia. Two of these stations including Chiang Mai (18.76°N, 98.93°E, dip angle 12.7°N) and Kototabang (0.2°S, 100.3°E, dip angle 10.1°S) are at the conjugate points while Chumphon (10.72°N, 99.37°E, dip angle 3.0°N) station is near the equator. To produce the model, the feed forward network with backpropagation algorithm is applied. The NN is trained with the daily hourly values of foF2 during 2004-2012, except 2009, and the selected input parameters, which affect the foF2 variability, include day number (DN), hour number (HR), solar zenith angle (C), geographic latitude (θ), magnetic inclination (I), magnetic declination (D) and angle of meridian (M) relative to the sub-solar point, the 7-day mean of F10.7 (F10.7_7), the 81-day mean of SSN (SSN_81) and the 2-day mean of Ap (Ap_2). The foF2 data of 2009 and 2013 are then used for testing the NN model during the foF2 interpolation and extrapolation, respectively. To examine the performance of the proposed NN, the root mean square error (RMSE) of the observed foF2, the proposed NN model and the IRI-2012 (CCIR and URSI options) model are compared. In general, the results show the same trends in foF2 variation between the models (NN and IRI-2012) and the observations in that they are higher during the day and lower at night. Besides, the results demonstrate that the proposed NN model can predict the foF2 values more closely during daytime than during nighttime as supported by the lower RMSE values during daytime (0.5 ≤ RMSE ≤ 1.0 for Chumphon and Kototabang, 0.7 ≤ RMSE ≤ 1.2 at Chiang Mai) and with the highest levels during nighttime (0.8 ≤ RMSE ≤ 1.5 for Chumphon and Kototabang, 1.2 ≤ RMSE ≤ 2.0 at Chiang Mai). Furthermore, the NN model predicts the foF2 values more accurately than the IRI model at the
EEG-fMRI Bayesian framework for neural activity estimation: a simulation study
Croce, Pierpaolo; Basti, Alessio; Marzetti, Laura; Zappasodi, Filippo; Del Gratta, Cosimo
2016-12-01
Objective. Due to the complementary nature of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI), and given the possibility of simultaneous acquisition, the joint data analysis can afford a better understanding of the underlying neural activity estimation. In this simulation study we want to show the benefit of the joint EEG-fMRI neural activity estimation in a Bayesian framework. Approach. We built a dynamic Bayesian framework in order to perform joint EEG-fMRI neural activity time course estimation. The neural activity is originated by a given brain area and detected by means of both measurement techniques. We have chosen a resting state neural activity situation to address the worst case in terms of the signal-to-noise ratio. To infer information by EEG and fMRI concurrently we used a tool belonging to the sequential Monte Carlo (SMC) methods: the particle filter (PF). Main results. First, despite a high computational cost, we showed the feasibility of such an approach. Second, we obtained an improvement in neural activity reconstruction when using both EEG and fMRI measurements. Significance. The proposed simulation shows the improvements in neural activity reconstruction with EEG-fMRI simultaneous data. The application of such an approach to real data allows a better comprehension of the neural dynamics.
Artificial neural network approach for estimation of surface specific ...
Indian Academy of Sciences (India)
Microwave sensor MSMR (Multifrequency Scanning Microwave Radiometer) data onboard Oceansat-1 was used for retrieval of monthly averages of near surface specific humidity (a) and air temperature (a) by means of Artificial Neural Network (ANN). The MSMR measures the microwave radiances in 8 channels at ...
Nonlinear neural network for hemodynamic model state and input estimation using fMRI data
Karam, Ayman M.
2014-11-01
Originally inspired by biological neural networks, artificial neural networks (ANNs) are powerful mathematical tools that can solve complex nonlinear problems such as filtering, classification, prediction and more. This paper demonstrates the first successful implementation of ANN, specifically nonlinear autoregressive with exogenous input (NARX) networks, to estimate the hemodynamic states and neural activity from simulated and measured real blood oxygenation level dependent (BOLD) signals. Blocked and event-related BOLD data are used to test the algorithm on real experiments. The proposed method is accurate and robust even in the presence of signal noise and it does not depend on sampling interval. Moreover, the structure of the NARX networks is optimized to yield the best estimate with minimal network architecture. The results of the estimated neural activity are also discussed in terms of their potential use.
H∞state estimation of stochastic memristor-based neural networks with time-varying delays.
Bao, Haibo; Cao, Jinde; Kurths, Jürgen; Alsaedi, Ahmed; Ahmad, Bashir
2018-03-01
This paper addresses the problem of H ∞ state estimation for a class of stochastic memristor-based neural networks with time-varying delays. Under the framework of Filippov solution, the stochastic memristor-based neural networks are transformed into systems with interval parameters. The present paper is the first to investigate the H ∞ state estimation problem for continuous-time Itô-type stochastic memristor-based neural networks. By means of Lyapunov functionals and some stochastic technique, sufficient conditions are derived to ensure that the estimation error system is asymptotically stable in the mean square with a prescribed H ∞ performance. An explicit expression of the state estimator gain is given in terms of linear matrix inequalities (LMIs). Compared with other results, our results reduce control gain and control cost effectively. Finally, numerical simulations are provided to demonstrate the efficiency of the theoretical results. Copyright © 2018 Elsevier Ltd. All rights reserved.
Application of radial basis neural network for state estimation of ...
African Journals Online (AJOL)
user
MultiCraft. International Journal of Engineering, Science and Technology. Vol. 2, No. 3, 2010, pp. 19-28. INTERNATIONAL. JOURNAL OF. ENGINEERING,. SCIENCE AND. TECHNOLOGY ... state estimation is investigated by testing its applicability on a IEEE 14 bus system. The proposed estimator is compared with.
Energy Technology Data Exchange (ETDEWEB)
Schuelke, J.S.; Quirein, J.A.; Sarg, J.F.
1998-12-31
This case study shows the benefit of using multiple seismic trace attributes and the pattern recognition capabilities of neural networks to predict reservoir architecture and porosity distribution in the Pegasus Field, West Texas. The study used the power of neural networks to integrate geologic, borehole and seismic data. Illustrated are the improvements between the new neural network approach and the more traditional method of seismic trace inversion for porosity estimation. Comprehensive statistical methods and interpretational/subjective measures are used in the prediction of porosity from seismic attributes. A 3-D volume of seismic derived porosity estimates for the Devonian reservoir provide a very detailed estimate of porosity, both spatially and vertically, for the field. The additional reservoir porosity detail provided, between the well control, allows for optimal placement of horizontal wells and improved field development. 6 refs., 2 figs.
Information-geometric measures estimate neural interactions during oscillatory brain states.
Nie, Yimin; Fellous, Jean-Marc; Tatsuno, Masami
2014-01-01
The characterization of functional network structures among multiple neurons is essential to understanding neural information processing. Information geometry (IG), a theory developed for investigating a space of probability distributions has recently been applied to spike-train analysis and has provided robust estimations of neural interactions. Although neural firing in the equilibrium state is often assumed in these studies, in reality, neural activity is non-stationary. The brain exhibits various oscillations depending on cognitive demands or when an animal is asleep. Therefore, the investigation of the IG measures during oscillatory network states is important for testing how the IG method can be applied to real neural data. Using model networks of binary neurons or more realistic spiking neurons, we studied how the single- and pairwise-IG measures were influenced by oscillatory neural activity. Two general oscillatory mechanisms, externally driven oscillations and internally induced oscillations, were considered. In both mechanisms, we found that the single-IG measure was linearly related to the magnitude of the external input, and that the pairwise-IG measure was linearly related to the sum of connection strengths between two neurons. We also observed that the pairwise-IG measure was not dependent on the oscillation frequency. These results are consistent with the previous findings that were obtained under the equilibrium conditions. Therefore, we demonstrate that the IG method provides useful insights into neural interactions under the oscillatory condition that can often be observed in the real brain.
Information-geometric measures estimate neural interactions during oscillatory brain states
Directory of Open Access Journals (Sweden)
Yimin eNie
2014-02-01
Full Text Available The characterization of functional network structures among multiple neurons is essential to understanding neural information processing. Information geometry (IG, a theory developed for investigating a space of probability distributions has recently been applied to spike-train analysis and has provided robust estimations of neural interactions. Although neural firing in the equilibrium state is often assumed in these studies, in reality, neural activity is non-stationary. The brain exhibits various oscillations depending on cognitive demands or when an animal is asleep. Therefore, the investigation of the IG measures during oscillatory network states is important for testing how the IG method can be applied to real neural data. Using model networks of binary neurons or more realistic spiking neurons, we studied how the single- and pairwise-IG measures were influenced by oscillatory neural activity. Two general oscillatory mechanisms, externally driven oscillations and internally induced oscillations, were considered. In both mechanisms, we found that the single-IG measure was linearly related to the magnitude of the external input, and that the pairwise-IG measure was linearly related to the sum of connection strengths between two neurons. We also observed that the pairwise-IG measure was not dependent on the oscillation frequency. These results are consistent with the previous findings that were obtained under the equilibrium conditions. Therefore, we demonstrate that the IG method provides useful insights into neural interactions under the oscillatory condition that can often be observed in the real brain.
Application of Artificial Neural Networks for Efficient High-Resolution 2D DOA Estimation
Directory of Open Access Journals (Sweden)
M. Agatonović
2012-12-01
Full Text Available A novel method to provide high-resolution Two-Dimensional Direction of Arrival (2D DOA estimation employing Artificial Neural Networks (ANNs is presented in this paper. The observed space is divided into azimuth and elevation sectors. Multilayer Perceptron (MLP neural networks are employed to detect the presence of a source in a sector while Radial Basis Function (RBF neural networks are utilized for DOA estimation. It is shown that a number of appropriately trained neural networks can be successfully used for the high-resolution DOA estimation of narrowband sources in both azimuth and elevation. The training time of each smaller network is significantly re¬duced as different training sets are used for networks in detection and estimation stage. By avoiding the spectral search, the proposed method is suitable for real-time ap¬plications as it provides DOA estimates in a matter of seconds. At the same time, it demonstrates the accuracy comparable to that of the super-resolution 2D MUSIC algorithm.
Directory of Open Access Journals (Sweden)
Víctor Martínez-Martínez
2013-06-01
Full Text Available This paper presents an Artificial Neural Network (ANN based model for environmental variables related to the tobacco drying process. A fitting ANN was used to estimate and predict temperature and relative humidity inside the tobacco dryer: the estimation consists of calculating the value of these variables in different locations of the dryer and the prediction consists of forecasting the value of these variables with different time horizons. The proposed model has been validated with temperature and relative humidity data obtained from a real tobacco dryer using a Wireless Sensor Network (WSN. On the one hand, an error under 2% was achieved, obtaining temperature as a function of temperature and relative humidity in other locations in the estimation task. Besides, an error around 1.5 times lower than the one obtained with an interpolation method was achieved in the prediction task when the temperature inside the tobacco mass was predicted with time horizons over 2.5 hours as a function of its present and past values. These results show that ANN-based models can be used to improve the tobacco drying process because with these types of models the value of environmental variables can be predicted in the near future and can be estimated in other locations with low errors.
Can shoulder joint reaction forces be estimated by neural networks?
de Vries, W.H.K.; Veeger, H.E.J.; Baten, C.T.M.; van der Helm, F.C.T.
2016-01-01
To facilitate the development of future shoulder endoprostheses, a long term load profile of the shoulder joint is desired. A musculoskeletal model using 3D kinematics and external forces as input can estimate the mechanical load on the glenohumeral joint, in terms of joint reaction forces. For long
Parameter estimation of breast tumour using dynamic neural network from thermal pattern
Directory of Open Access Journals (Sweden)
Elham Saniei
2016-11-01
Full Text Available This article presents a new approach for estimating the depth, size, and metabolic heat generation rate of a tumour. For this purpose, the surface temperature distribution of a breast thermal image and the dynamic neural network was used. The research consisted of two steps: forward and inverse. For the forward section, a finite element model was created. The Pennes bio-heat equation was solved to find surface and depth temperature distributions. Data from the analysis, then, were used to train the dynamic neural network model (DNN. Results from the DNN training/testing confirmed those of the finite element model. For the inverse section, the trained neural network was applied to estimate the depth temperature distribution (tumour position from the surface temperature profile, extracted from the thermal image. Finally, tumour parameters were obtained from the depth temperature distribution. Experimental findings (20 patients were promising in terms of the model’s potential for retrieving tumour parameters.
Simulation, State Estimation and Control of Nonlinear Superheater Attemporator using Neural Networks
DEFF Research Database (Denmark)
Bendtsen, Jan Dimon; Sørensen, O.
2000-01-01
This paper considers the use of neural networks for nonlinear state estimation, system identification and control. As a case study we use data taken from a nonlinear injection valve for a superheater attemporator at a power plant. One neural network is trained as a nonlinear simulation model......-by-sample linearizations and state estimates provided by the observer network. Simulation studies show that the nonlinear observer-based control loop performs better than a similar control loop based on a linear observer....... of the process, then another network is trained to act as a combined state and parameter estimator for the process. The observer network incorporates smoothing of the parameter estimates in the form of regularization. A pole placement controller is designed which takes advantage of the sample...
Simulation, State Estimation and Control of Nonlinear Superheater Attemporator using Neural Networks
DEFF Research Database (Denmark)
Bendtsen, Jan Dimon; Sørensen, O.
1999-01-01
This paper considers the use of neural networks for nonlinear state estimation, system identification and control. As a case study we use data taken from a nonlinear injection valve for a superheater attemporator at a power plant. One neural network is trained as a nonlinear simulation model......-by-sample linearizations and state estimates provided by the observer network. Simulation studies show that the nonlinear observer-based control loop performs better than a similar control loop based on a linear observer....... of the process, then another network is trained to act as a combined state and parameter estimator for the process. The observer network incorporates smoothing of the parameter estimates in the form of regularization. A pole placement controller is designed which takes advantage of the sample...
A dynamic programming approach to missing data estimation using neural networks
CSIR Research Space (South Africa)
Nelwamondo, FV
2013-01-01
Full Text Available This paper develops and presents a novel technique for missing data estimation using a combination of dynamic programming, neural networks and genetic algorithms (GA) on suitable subsets of the input data. The method proposed here is well suited...
Neural estimation of kinetic rate constants from dynamic PET-scans
DEFF Research Database (Denmark)
Fog, Torben L.; Nielsen, Lars Hupfeldt; Hansen, Lars Kai
1994-01-01
A feedforward neural net is trained to invert a simple three compartment model describing the tracer kinetics involved in the metabolism of [18F]fluorodeoxyglucose in the human brain. The network can estimate rate constants from positron emission tomography sequences and is about 50 times faster...... than direct fitting of rate constants using the parametrized transients of the compartment model...
Witoonchart, Peerajak; Chongstitvatana, Prabhas
2017-08-01
In this study, for the first time, we show how to formulate a structured support vector machine (SSVM) as two layers in a convolutional neural network, where the top layer is a loss augmented inference layer and the bottom layer is the normal convolutional layer. We show that a deformable part model can be learned with the proposed structured SVM neural network by backpropagating the error of the deformable part model to the convolutional neural network. The forward propagation calculates the loss augmented inference and the backpropagation calculates the gradient from the loss augmented inference layer to the convolutional layer. Thus, we obtain a new type of convolutional neural network called an Structured SVM convolutional neural network, which we applied to the human pose estimation problem. This new neural network can be used as the final layers in deep learning. Our method jointly learns the structural model parameters and the appearance model parameters. We implemented our method as a new layer in the existing Caffe library. Copyright © 2017 Elsevier Ltd. All rights reserved.
Directory of Open Access Journals (Sweden)
Ahmad Aryafar
2016-06-01
Full Text Available Nowadays, estimating the ampere consumption and achieve to the optimum condition from the perspective of energy consumption is one of the most important steps to reduce the production costs. In this research it is tried to develop an accurate model for estimating the ampere consumption by using the artificial neural networks (ANN.In the first step, experimental studies were carried out on 7 carbonate rock samples in different conditions at particular feed rates (100, 200, 300and 400 and depth of cuts (15, 22, 30and 35mm using a fully instrumented laboratory rig that is enable to change the machine parameters and measure the ampere consumption. In next step, a back propagation neural network was designed for modelling the sawing process for predicting the ampere consumption. The input network consisting of two parts: machine, work piece characteristics and the output of neural network was ampere consumption. This research evaluated the competencies of neural networks to estimate the ampere consumption in sawing process. The correlation coefficient between measured and predicted data in training and testing data is 0.95 and 0.97 respectively. The root mean square error (RMSE for train and test data is 1.2 and 0.7 respectively. The results of this study showed that the ANNs can be used to estimate the ampere consumption with high ability and low error for industrial applications. Moreover, the cost of sawing machine ampere consumption can be accurately estimated using this neural model from some important physical and mechanical properties of rock.
MEG and fMRI fusion for nonlinear estimation of neural and BOLD signal changes
Directory of Open Access Journals (Sweden)
Sergey M Plis
2010-11-01
Full Text Available The combined analysis of MEG/EEG and functional MRI measurements can lead to improvement in the description of the dynamical and spatial properties of brain activity. In this paper we empirically demonstrate this improvement using simulated and recorded task related MEG and fMRI activity. Neural activity estimates were derived using a dynamic Bayesian network with continuous real valued parameters by means of a sequential Monte Carlo technique. In synthetic data, we show that MEG and fMRI fusion improves estimation of the indirectly observed neural activity and smooths tracking of the BOLD response. In recordings of task related neural activity the combination of MEG and fMRI produces a result with greater SNR, that confirms the expectation arising from the nature of the experiment. The highly nonlinear model of the BOLD response poses a difficult inference problem for neural activity estimation; computational requirements are also high due to the time and space complexity. We show that joint analysis of the data improves the system's behavior by stabilizing the differential equations system and by requiring fewer computational resources.
Directory of Open Access Journals (Sweden)
Dawei Han
2012-02-01
Full Text Available The application of ANNs (Artifi cial Neural Networks has been studied by many researchers in modelling rainfall runoff processes. However, the work so far has been focused on the rainfall data from traditional raingauges. Weather radar is a modern technology which could provide high resolution rainfall in time and space. In this study, a comparison in rainfall runoff modelling between the raingauge and weather radar has been carried out. The data were collected from Brue catchment in Southwest of England, with 49 raingauges covering 136 km2 and two C-band weather radars. This raingauge network is extremely dense (for research purposes and does not represent the usual raingauge density in operational flood forecasting systems. The ANN models were set up with both lumped and spatial rainfall input. The results showed that raingauge data outperformed radar data in all the events tested, regardless of the lumped and spatial input. La aplicación de Redes Neuronales Artificiales (RNA en el modelado de lluvia-flujo ha sido estudiada ampliamente. Sin embargo, hasta ahora se han utilizado datos provenientes de pluviómetros tradicionales. Los radares meteorológicos son una tecnología moderna que puede proveer datos de lluvia de alta resolución en tiempo y espacio. Este es un trabajo de comparación en el modelado lluvia-flujo entre pluviómetros y radares meteorológicos. Los datos provienen de la cuenca del río Brue en el suroeste de Inglaterra, con 49 pluviómetros cubriendo 136 km2 y dos radares meteorológicos en la banda C. Esta red de pluviómetros es extremadamente densa (para investigación y no representa la densidad usual en sistemas de predicción de inundaciones. Los modelos de RNA fueron implementados con datos de entrada de lluvia tanto espaciados como no distribuidos. Los resultados muestran que los datos de los pluviómetros fueron mejores que los datos de los radares en todos los eventos probados.
Reactor vessel water level estimation during severe accidents using cascaded fuzzy neural networks
Energy Technology Data Exchange (ETDEWEB)
Kim, Dong Yeong; Yoo, Kwae Hwan; Choi, Geon Pil; Back, Ju Hyun; Na, Man Gyun [Dept. of Nuclear Engineering, Chosun University, Gwangju (Korea, Republic of)
2016-06-15
Global concern and interest in the safety of nuclear power plants have increased considerably since the Fukushima accident. In the event of a severe accident, the reactor vessel water level cannot be measured. The reactor vessel water level has a direct impact on confirming the safety of reactor core cooling. However, in the event of a severe accident, it may be possible to estimate the reactor vessel water level by employing other information. The cascaded fuzzy neural network (CFNN) model can be used to estimate the reactor vessel water level through the process of repeatedly adding fuzzy neural networks. The developed CFNN model was found to be sufficiently accurate for estimating the reactor vessel water level when the sensor performance had deteriorated. Therefore, the developed CFNN model can help provide effective information to operators in the event of a severe accident.
Directory of Open Access Journals (Sweden)
Weibo Zhao
2017-12-01
Full Text Available Power generation industry is the key industry of carbon dioxide (CO2 emission in China. Assessing its future CO2 emissions is of great significance to the formulation and implementation of energy saving and emission reduction policies. Based on the Stochastic Impacts by Regression on Population, Affluence and Technology model (STIRPAT, the influencing factors analysis model of CO2 emission of power generation industry is established. The ridge regression (RR method is used to estimate the historical data. In addition, a wavelet neural network (WNN prediction model based on Cuckoo Search algorithm optimized by Gauss (GCS is put forward to predict the factors in the STIRPAT model. Then, the predicted values are substituted into the regression model, and the CO2 emission estimation values of the power generation industry in China are obtained. It’s concluded that population, per capita Gross Domestic Product (GDP, standard coal consumption and thermal power specific gravity are the key factors affecting the CO2 emission from the power generation industry. Besides, the GCS-WNN prediction model has higher prediction accuracy, comparing with other models. Moreover, with the development of science and technology in the future, the CO2 emission growth in the power generation industry will gradually slow down according to the prediction results.
Wang, Y; Zhu, W; Cheng, X; Li, D
2013-03-07
Continuous crystal based PET detectors have features of simple design, low cost, good energy resolution and high detection efficiency. Through single-end readout of scintillation light, direct three-dimensional (3D) position estimation could be another advantage that the continuous crystal detector would have. In this paper, we propose to use artificial neural networks to simultaneously estimate the plane coordinate and DOI coordinate of incident γ photons with detected scintillation light. Using our experimental setup with an '8 + 8' simplified signal readout scheme, the training data of perpendicular irradiation on the front surface and one side surface are obtained, and the plane (x, y) networks and DOI networks are trained and evaluated. The test results show that the artificial neural network for DOI estimation is as effective as for plane estimation. The performance of both estimators is presented by resolution and bias. Without bias correction, the resolution of the plane estimator is on average better than 2 mm and that of the DOI estimator is about 2 mm over the whole area of the detector. With bias correction, the resolution at the edge area for plane estimation or at the end of the block away from the readout PMT for DOI estimation becomes worse, as we expect. The comprehensive performance of the 3D positioning by a neural network is accessed by the experimental test data of oblique irradiations. To show the combined effect of the 3D positioning over the whole area of the detector, the 2D flood images of oblique irradiation are presented with and without bias correction.
Energy Technology Data Exchange (ETDEWEB)
Koh, Hee Seok; Kim, Hyun Duck [Kyungnam University, Masan (Korea, Republic of); Lee, Kwang Seok [Chinju National University (Korea, Republic of)
1996-02-01
In this paper, we proposed an estimation method of a posterior probability and PDF(Probability density function) using a feed forward neural network and code books of VQ(vector quantization). In this study, We estimates a posterior probability and probability density function, which compose a new parameter with well-known Mel cepstrum and verificate the performance for the five vowels taking from syllables by NN(neural network) and PNN(probabilistic neural network). In case of new parameter, showed the best result by probabilistic neural network and recognition rates are average 83.02%. (author). 7 refs., 4 figs., 3 tabs.
Directory of Open Access Journals (Sweden)
Pengshan Li
2016-10-01
Full Text Available With the continued social and economic development of northern China, landscape fragmentation has placed increasing pressure on the ecological system of the Beijing-Tianjin-Hebei (BTH region. To maintain the integrity of ecological processes under the influence of human activities, we must maintain effective connections between habitats and limit the impact of ecological isolation. In this paper, landscape elements were identified based on a kernel density estimation, including forests, grasslands, orchards and wetlands. The spatial configuration of ecological networks was analysed by the integrated density index, and a natural breaks classification was performed for the landscape type data and the results of the landscape spatial distribution analysis. The results showed that forest and grassland are the primary constituents of the core areas and act as buffer zones for the region’s ecological network. Rivers, as linear patches, and orchards, as stepping stones, form the main body of the ecological corridors, and isolated elements are distributed mainly in the plain area. Orchards have transition effects. Wetlands act as connections between different landscapes in the region. Based on these results, we make suggestions for the protection and planning of ecological networks. This study can also provide guidance for the coordinated development of the BTH region.
Xia, Peng; Hu, Jie; Peng, Yinghong
2017-10-25
A novel model based on deep learning is proposed to estimate kinematic information for myoelectric control from multi-channel electromyogram (EMG) signals. The neural information of limb movement is embedded in EMG signals that are influenced by all kinds of factors. In order to overcome the negative effects of variability in signals, the proposed model employs the deep architecture combining convolutional neural networks (CNNs) and recurrent neural networks (RNNs). The EMG signals are transformed to time-frequency frames as the input to the model. The limb movement is estimated by the model that is trained with the gradient descent and backpropagation procedure. We tested the model for simultaneous and proportional estimation of limb movement in eight healthy subjects and compared it with support vector regression (SVR) and CNNs on the same data set. The experimental studies show that the proposed model has higher estimation accuracy and better robustness with respect to time. The combination of CNNs and RNNs can improve the model performance compared with using CNNs alone. The model of deep architecture is promising in EMG decoding and optimization of network structures can increase the accuracy and robustness. © 2017 International Center for Artificial Organs and Transplantation and Wiley Periodicals, Inc.
Directory of Open Access Journals (Sweden)
Haorui Liu
2016-01-01
Full Text Available In the car control systems, it is hard to measure some key vehicle states directly and accurately when running on the road and the cost of the measurement is high as well. To address these problems, a vehicle state estimation method based on the kernel principal component analysis and the improved Elman neural network is proposed. Combining with nonlinear vehicle model of three degrees of freedom (3 DOF, longitudinal, lateral, and yaw motion, this paper applies the method to the soft sensor of the vehicle states. The simulation results of the double lane change tested by Matlab/SIMULINK cosimulation prove the KPCA-IENN algorithm (kernel principal component algorithm and improved Elman neural network to be quick and precise when tracking the vehicle states within the nonlinear area. This algorithm method can meet the software performance requirements of the vehicle states estimation in precision, tracking speed, noise suppression, and other aspects.
Synchronization-based parameter estimation of fractional-order neural networks
Gu, Yajuan; Yu, Yongguang; Wang, Hu
2017-10-01
This paper focuses on the parameter estimation problem of fractional-order neural network. By combining the adaptive control and parameter update law, we generalize the synchronization-based identification method that has been reported in several literatures on identifying unknown parameters of integer-order systems. With this method, parameter identification and synchronization can be achieved simultaneously. Finally, a numerical example is given to illustrate the effectiveness of the theoretical results.
DEFF Research Database (Denmark)
Soliman, Hammam Abdelaal Hammam; Wang, Huai; Gadalla, Brwene Salah Abdelkarim
2015-01-01
challenges. A capacitance estimation method based on Artificial Neural Network (ANN) algorithm is therefore proposed in this paper. The implemented ANN estimated the capacitance of the DC-link capacitor in a back-toback converter. Analysis of the error of the capacitance estimation is also given......In power electronic converters, reliability of DC-link capacitors is one of the critical issues. The estimation of their health status as an application of condition monitoring have been an attractive subject for industrial field and hence for the academic research filed as well. More reliable...... solutions are required to be adopted by the industry applications in which usage of extra hardware, increased cost, and low estimation accuracy are the main challenges. Therefore, development of new condition monitoring methods based on software solutions could be the new era that covers the aforementioned...
Uncertainties in neural network model based on carbon dioxide concentration for occupancy estimation
Energy Technology Data Exchange (ETDEWEB)
Alam, Azimil Gani; Rahman, Haolia; Kim, Jung-Kyung; Han, Hwataik [Kookmin University, Seoul (Korea, Republic of)
2017-05-15
Demand control ventilation is employed to save energy by adjusting airflow rate according to the ventilation load of a building. This paper investigates a method for occupancy estimation by using a dynamic neural network model based on carbon dioxide concentration in an occupied zone. The method can be applied to most commercial and residential buildings where human effluents to be ventilated. An indoor simulation program CONTAMW is used to generate indoor CO{sub 2} data corresponding to various occupancy schedules and airflow patterns to train neural network models. Coefficients of variation are obtained depending on the complexities of the physical parameters as well as the system parameters of neural networks, such as the numbers of hidden neurons and tapped delay lines. We intend to identify the uncertainties caused by the model parameters themselves, by excluding uncertainties in input data inherent in measurement. Our results show estimation accuracy is highly influenced by the frequency of occupancy variation but not significantly influenced by fluctuation in the airflow rate. Furthermore, we discuss the applicability and validity of the present method based on passive environmental conditions for estimating occupancy in a room from the viewpoint of demand control ventilation applications.
Perreault Levasseur, Laurence; Hezaveh, Yashar D.; Wechsler, Risa H.
2017-11-01
In Hezaveh et al. we showed that deep learning can be used for model parameter estimation and trained convolutional neural networks to determine the parameters of strong gravitational-lensing systems. Here we demonstrate a method for obtaining the uncertainties of these parameters. We review the framework of variational inference to obtain approximate posteriors of Bayesian neural networks and apply it to a network trained to estimate the parameters of the Singular Isothermal Ellipsoid plus external shear and total flux magnification. We show that the method can capture the uncertainties due to different levels of noise in the input data, as well as training and architecture-related errors made by the network. To evaluate the accuracy of the resulting uncertainties, we calculate the coverage probabilities of marginalized distributions for each lensing parameter. By tuning a single variational parameter, the dropout rate, we obtain coverage probabilities approximately equal to the confidence levels for which they were calculated, resulting in accurate and precise uncertainty estimates. Our results suggest that the application of approximate Bayesian neural networks to astrophysical modeling problems can be a fast alternative to Monte Carlo Markov Chains, allowing orders of magnitude improvement in speed.
Gross domestic product estimation based on electricity utilization by artificial neural network
Stevanović, Mirjana; Vujičić, Slađana; Gajić, Aleksandar M.
2018-01-01
The main goal of the paper was to estimate gross domestic product (GDP) based on electricity estimation by artificial neural network (ANN). The electricity utilization was analyzed based on different sources like renewable, coal and nuclear sources. The ANN network was trained with two training algorithms namely extreme learning method and back-propagation algorithm in order to produce the best prediction results of the GDP. According to the results it can be concluded that the ANN model with extreme learning method could produce the acceptable prediction of the GDP based on the electricity utilization.
Alternative Sensor System and MLP Neural Network for Vehicle Pedal Activity Estimation
Directory of Open Access Journals (Sweden)
Ahmed M. Wefky
2010-04-01
Full Text Available It is accepted that the activity of the vehicle pedals (i.e., throttle, brake, clutch reflects the driver’s behavior, which is at least partially related to the fuel consumption and vehicle pollutant emissions. This paper presents a solution to estimate the driver activity regardless of the type, model, and year of fabrication of the vehicle. The solution is based on an alternative sensor system (regime engine, vehicle speed, frontal inclination and linear acceleration that reflects the activity of the pedals in an indirect way, to estimate that activity by means of a multilayer perceptron neural network with a single hidden layer.
Miao, Zhiyong; Shi, Hongyang; Zhang, Yi; Xu, Fan
2017-10-01
In this paper, a new variational Bayesian adaptive cubature Kalman filter (VBACKF) is proposed for nonlinear state estimation. Although the conventional VBACKF performs better than cubature Kalman filtering (CKF) in solving nonlinear systems with time-varying measurement noise, its performance may degrade due to the uncertainty of the system model. To overcome this drawback, a multilayer feed-forward neural network (MFNN) is used to aid the conventional VBACKF, generalizing it to attain higher estimation accuracy and robustness. In the proposed neural-network-aided variational Bayesian adaptive cubature Kalman filter (NN-VBACKF), the MFNN is used to turn the state estimation of the VBACKF adaptively, and it is used for both state estimation and in the online training paradigm simultaneously. To evaluate the performance of the proposed method, it is compared with CKF and VBACKF via target tracking problems. The simulation results demonstrate that the estimation accuracy and robustness of the proposed method are better than those of the CKF and VBACKF.
Comparison of Artificial Neural Networks and GIS Based Solar Analysis for Solar Potential Estimation
Konakoǧlu, Berkant; Usta, Ziya; Cömert, Çetin; Gökalp, Ertan
2016-04-01
Nowadays, estimation of solar potential plays an important role in planning process for sustainable cities. The use of solar panels, which produces electricity directly from the sun, has become popular in accordance with developing technologies. Since the use of solar panels enables the users to decrease costs and increase yields, the use of solar panels will be more popular in the future. Production of electricity is not convenient for all circumstances. Shading effects, massive clouds and rainy weather are some factors that directly affect the production of electricity from solar energy. Hence, before the installation of solar panels, it is crucial to conduct spatial analysis and estimate the solar potential of the place that the solar panel will be installed. There are several approaches to determine the solar potential. Examination of the applications in the literature reveals that the applications conducted for determining the solar potential are divided into two main categories. Solar potential is estimated either by using artificial neural network approach in which statistical parameters such as the duration of sun shine, number of clear days, solar radiation etc. are used, or by spatial analysis conducted in GIS approaches in which spatial parameters such as, latitude, longitude, slope, aspect etc. are used. In the literature, there are several studies that use both approaches but the literature lacks of a study related to the comparison of these approaches. In this study, Karadeniz Technical University campus has been selected as study area. Monthly average values of the number of clear sky days, air temperature, atmospheric pressure, relative humidity, sunshine duration and solar radiation parameters obtained for the years between 2005 and 2015 will be used to perform artificial neural network analysis to estimate the solar potential of the study area. The solar potential will also be estimated by using GIS-based solar analysis modules. The results of
The use of neural networks in identifying error sources in satellite-derived tropical SST estimates.
Lee, Yung-Hsiang; Ho, Chung-Ru; Su, Feng-Chun; Kuo, Nan-Jung; Cheng, Yu-Hsin
2011-01-01
An neural network model of data mining is used to identify error sources in satellite-derived tropical sea surface temperature (SST) estimates from thermal infrared sensors onboard the Geostationary Operational Environmental Satellite (GOES). By using the Back Propagation Network (BPN) algorithm, it is found that air temperature, relative humidity, and wind speed variation are the major factors causing the errors of GOES SST products in the tropical Pacific. The accuracy of SST estimates is also improved by the model. The root mean square error (RMSE) for the daily SST estimate is reduced from 0.58 K to 0.38 K and mean absolute percentage error (MAPE) is 1.03%. For the hourly mean SST estimate, its RMSE is also reduced from 0.66 K to 0.44 K and the MAPE is 1.3%.
The Use of Neural Networks in Identifying Error Sources in Satellite-Derived Tropical SST Estimates
Directory of Open Access Journals (Sweden)
Yu-Hsin Cheng
2011-07-01
Full Text Available An neural network model of data mining is used to identify error sources in satellite-derived tropical sea surface temperature (SST estimates from thermal infrared sensors onboard the Geostationary Operational Environmental Satellite (GOES. By using the Back Propagation Network (BPN algorithm, it is found that air temperature, relative humidity, and wind speed variation are the major factors causing the errors of GOES SST products in the tropical Pacific. The accuracy of SST estimates is also improved by the model. The root mean square error (RMSE for the daily SST estimate is reduced from 0.58 K to 0.38 K and mean absolute percentage error (MAPE is 1.03%. For the hourly mean SST estimate, its RMSE is also reduced from 0.66 K to 0.44 K and the MAPE is 1.3%.
Firing Rate Estimation Using Infinite Mixture Models and Its Application to Neural Decoding.
Shibue, Ryohei; Komaki, Fumiyasu
2017-08-09
Neural decoding is a framework for reconstructing external stimuli from spike trains recorded in brains. Kloosterman et al. (2014) proposed a new decoding method using marked point processes. This method does not require spike sorting and thereby improves decoding accuracy dramatically. In this method, they used kernel density estimation to estimate intensity functions of marked point processes. However, using kernel density estimation causes problems. To overcome these problems, we propose a new decoding method using infinite mixture models to estimate intensity. The proposed method improves decoding performance in terms of accuracy and computation speed. We apply the proposed method to simulation and experimental data to verify its performance. Copyright © 2016, Journal of Neurophysiology.
Lee, Jong-Ha; Won, Chang-Hee
2011-01-01
In this paper, we developed a methodology for estimating three parameters of tissue inclusion: size, depth, and Young's modulus from the tactile data obtained at the tissue surface with the tactile sensation imaging system. The estimation method consists of the forward algorithm using finite element method, and inversion algorithm using artificial neural network. The forward algorithm is designed to comprehensively predict the tactile data based on the mechanical properties of the tissue inclusion. This forward information is used to develop an inversion algorithm that will be used to extract the size, depth, and Young's modulus of a tissue inclusion from the tactile image. The proposed method is then validated with custom made tissue phantoms with matching elasticities of typical human breast tissues. The experimental results showed that the proposed estimation method estimates the size, depth, and Young's modulus of tissue inclusions with root mean squared errors of 1.25 mm, 2.09 mm, and 28.65 kPa, respectively.
Hammerstein-Wiener Model: A New Approach to the Estimation of Formal Neural Information
Directory of Open Access Journals (Sweden)
Reza Abbasi-Asl
2012-09-01
Full Text Available A new approach is introduced to estimate the formal information of neurons. Formal Information, mainly discusses about the aspects of the response that is related to the stimulus. Estimation is based on introducing a mathematical nonlinear model with Hammerstein-Wiener system estimator. This method of system identification consists of three blocks to completely describe the nonlinearity of input and output and linear behaviour of the model. The introduced model is trained by 166 spikes of neurons and other 166 spikes are used to test and validate the model. The simulation results show the R-Value of 92.6 % between estimated and reference information rate. This shows improvement of 1.41 % in comparison with MLP neural network.
A new EC-PC threshold estimation method for in vivo neural spike detection
Yang, Zhi; Liu, Wentai; Keshtkaran, Mohammad Reza; Zhou, Yin; Xu, Jian; Pikov, Victor; Guan, Cuntai; Lian, Yong
2012-08-01
This paper models in vivo neural signals and noise for extracellular spike detection. Although the recorded data approximately follow Gaussian distribution, they clearly deviate from white Gaussian noise due to neuronal synchronization and sparse distribution of spike energy. Our study predicts the coexistence of two components embedded in neural data dynamics, one in the exponential form (noise) and the other in the power form (neural spikes). The prediction of the two components has been confirmed in experiments of in vivo sequences recorded from the hippocampus, cortex surface, and spinal cord; both acute and long-term recordings; and sleep and awake states. These two components are further used as references for threshold estimation. Different from the conventional wisdom of setting a threshold at 3×RMS, the estimated threshold exhibits a significant variation. When our algorithm was tested on synthesized sequences with a different signal to noise ratio and on/off firing dynamics, inferred threshold statistics track the benchmarks well. We envision that this work may be applied to a wide range of experiments as a front-end data analysis tool.
High-precision position estimation in PET using artificial neural networks
Energy Technology Data Exchange (ETDEWEB)
Mateo, F. [Digital Systems Design Group (DSD), Instituto de las Tecnologias de la Informacion y de las Comunicaciones Avanzadas (ITACA), Universidad Politecnica de Valencia, Camino de Vera s/n, 46022 Valencia (Spain)], E-mail: fermaji@upvnet.upv.es; Aliaga, R.J.; Ferrando, N.; Martinez, J.D.; Herrero, V.; Lerche, Ch.W.; Colom, R.J.; Monzo, J.M.; Sebastia, A.; Gadea, R. [Digital Systems Design Group (DSD), Instituto de las Tecnologias de la Informacion y de las Comunicaciones Avanzadas (ITACA), Universidad Politecnica de Valencia, Camino de Vera s/n, 46022 Valencia (Spain)
2009-06-01
Traditionally, the most popular technique to predict the impact position of gamma photons on a PET detector has been Anger's logic. However, it introduces nonlinearities that compress the light distribution, reducing the useful field of view and the spatial resolution, especially at the edges of the scintillator crystal. In this work, we make use of neural networks to address a bias-corrected position estimation from real stimulus obtained from a 2D PET system setup. The preprocessing and data acquisition were performed by separate custom boards, especially designed for this application. The results show that neural networks yield a more uniform field of view while improving the systematic error and the spatial resolution. Therefore, they stand as a better performing and readily available alternative to classic positioning methods.
Estimation for Up/Down Fluctuation of Stock Prices by Using Neural Network
Watanabe, Toyohide; Iwata, Kenji
In general, it is not always easy to estimate stock prices exactly and get profits. Until today, many researchers have attacked to this subject, but could not report the successful estimation methods even if various approaches or many heuristics were applied in our knowledge-oriented society. This is because the fluctuation of stock prices is inherently characterized as random walk. In this paper, we address a short-term-specific up/down fluctuation estimation method of stock prices. Our approach is first to select 16 brand companies in Japan Stock Market as the fundamental stock features, and then to define analytically 8 stock attributes as input parameters for our 3-level neural network. We used 32,000 samples of 2,000 days from 16 brands: the first 1,000 days samples were used as learning data for our neural network; and the last 1,000 days samples were as test data. Our experiments showed that the up/down fluctuation estimation method in the short-term from the end value of today to the start value of tomorrow functions effectively.
Directory of Open Access Journals (Sweden)
Jinjun Tang
Full Text Available Travel time is an important measurement used to evaluate the extent of congestion within road networks. This paper presents a new method to estimate the travel time based on an evolving fuzzy neural inference system. The input variables in the system are traffic flow data (volume, occupancy, and speed collected from loop detectors located at points both upstream and downstream of a given link, and the output variable is the link travel time. A first order Takagi-Sugeno fuzzy rule set is used to complete the inference. For training the evolving fuzzy neural network (EFNN, two learning processes are proposed: (1 a K-means method is employed to partition input samples into different clusters, and a Gaussian fuzzy membership function is designed for each cluster to measure the membership degree of samples to the cluster centers. As the number of input samples increases, the cluster centers are modified and membership functions are also updated; (2 a weighted recursive least squares estimator is used to optimize the parameters of the linear functions in the Takagi-Sugeno type fuzzy rules. Testing datasets consisting of actual and simulated data are used to test the proposed method. Three common criteria including mean absolute error (MAE, root mean square error (RMSE, and mean absolute relative error (MARE are utilized to evaluate the estimation performance. Estimation results demonstrate the accuracy and effectiveness of the EFNN method through comparison with existing methods including: multiple linear regression (MLR, instantaneous model (IM, linear model (LM, neural network (NN, and cumulative plots (CP.
Improved Shape Parameter Estimation in Pareto Distributed Clutter with Neural Networks
Directory of Open Access Journals (Sweden)
José Raúl Machado-Fernández
2016-12-01
Full Text Available The main problem faced by naval radars is the elimination of the clutter input which is a distortion signal appearing mixed with target reflections. Recently, the Pareto distribution has been related to sea clutter measurements suggesting that it may provide a better fit than other traditional distributions. The authors propose a new method for estimating the Pareto shape parameter based on artificial neural networks. The solution achieves a precise estimation of the parameter, having a low computational cost, and outperforming the classic method which uses Maximum Likelihood Estimates (MLE. The presented scheme contributes to the development of the NATE detector for Pareto clutter, which uses the knowledge of clutter statistics for improving the stability of the detection, among other applications.
Sedaghat, A.; Bayat, H.; Safari Sinegani, A. A.
2016-03-01
The saturated hydraulic conductivity ( K s ) of the soil is one of the main soil physical properties. Indirect estimation of this parameter using pedo-transfer functions (PTFs) has received considerable attention. The Purpose of this study was to improve the estimation of K s using fractal parameters of particle and micro-aggregate size distributions in smectitic soils. In this study 260 disturbed and undisturbed soil samples were collected from Guilan province, the north of Iran. The fractal model of Bird and Perrier was used to compute the fractal parameters of particle and micro-aggregate size distributions. The PTFs were developed by artificial neural networks (ANNs) ensemble to estimate K s by using available soil data and fractal parameters. There were found significant correlations between K s and fractal parameters of particles and microaggregates. Estimation of K s was improved significantly by using fractal parameters of soil micro-aggregates as predictors. But using geometric mean and geometric standard deviation of particles diameter did not improve K s estimations significantly. Using fractal parameters of particles and micro-aggregates simultaneously, had the most effect in the estimation of K s . Generally, fractal parameters can be successfully used as input parameters to improve the estimation of K s in the PTFs in smectitic soils. As a result, ANNs ensemble successfully correlated the fractal parameters of particles and micro-aggregates to K s .
Arabzadeh, Vida; Niaki, S. T. A.; Arabzadeh, Vahid
2017-10-01
One of the most important processes in the early stages of construction projects is to estimate the cost involved. This process involves a wide range of uncertainties, which make it a challenging task. Because of unknown issues, using the experience of the experts or looking for similar cases are the conventional methods to deal with cost estimation. The current study presents data-driven methods for cost estimation based on the application of artificial neural network (ANN) and regression models. The learning algorithms of the ANN are the Levenberg-Marquardt and the Bayesian regulated. Moreover, regression models are hybridized with a genetic algorithm to obtain better estimates of the coefficients. The methods are applied in a real case, where the input parameters of the models are assigned based on the key issues involved in a spherical tank construction. The results reveal that while a high correlation between the estimated cost and the real cost exists; both ANNs could perform better than the hybridized regression models. In addition, the ANN with the Levenberg-Marquardt learning algorithm (LMNN) obtains a better estimation than the ANN with the Bayesian-regulated learning algorithm (BRNN). The correlation between real data and estimated values is over 90%, while the mean square error is achieved around 0.4. The proposed LMNN model can be effective to reduce uncertainty and complexity in the early stages of the construction project.
Manifold absolute pressure estimation using neural network with hybrid training algorithm.
Muslim, Mohd Taufiq; Selamat, Hazlina; Alimin, Ahmad Jais; Haniff, Mohamad Fadzli
2017-01-01
In a modern small gasoline engine fuel injection system, the load of the engine is estimated based on the measurement of the manifold absolute pressure (MAP) sensor, which took place in the intake manifold. This paper present a more economical approach on estimating the MAP by using only the measurements of the throttle position and engine speed, resulting in lower implementation cost. The estimation was done via two-stage multilayer feed-forward neural network by combining Levenberg-Marquardt (LM) algorithm, Bayesian Regularization (BR) algorithm and Particle Swarm Optimization (PSO) algorithm. Based on the results found in 20 runs, the second variant of the hybrid algorithm yields a better network performance than the first variant of hybrid algorithm, LM, LM with BR and PSO by estimating the MAP closely to the simulated MAP values. By using a valid experimental training data, the estimator network that trained with the second variant of the hybrid algorithm showed the best performance among other algorithms when used in an actual retrofit fuel injection system (RFIS). The performance of the estimator was also validated in steady-state and transient condition by showing a closer MAP estimation to the actual value.
Manifold absolute pressure estimation using neural network with hybrid training algorithm.
Directory of Open Access Journals (Sweden)
Mohd Taufiq Muslim
Full Text Available In a modern small gasoline engine fuel injection system, the load of the engine is estimated based on the measurement of the manifold absolute pressure (MAP sensor, which took place in the intake manifold. This paper present a more economical approach on estimating the MAP by using only the measurements of the throttle position and engine speed, resulting in lower implementation cost. The estimation was done via two-stage multilayer feed-forward neural network by combining Levenberg-Marquardt (LM algorithm, Bayesian Regularization (BR algorithm and Particle Swarm Optimization (PSO algorithm. Based on the results found in 20 runs, the second variant of the hybrid algorithm yields a better network performance than the first variant of hybrid algorithm, LM, LM with BR and PSO by estimating the MAP closely to the simulated MAP values. By using a valid experimental training data, the estimator network that trained with the second variant of the hybrid algorithm showed the best performance among other algorithms when used in an actual retrofit fuel injection system (RFIS. The performance of the estimator was also validated in steady-state and transient condition by showing a closer MAP estimation to the actual value.
Automatic Estimation of the Dynamics of Channel Conductance Using a Recurrent Neural Network
Directory of Open Access Journals (Sweden)
Masaaki Takahashi
2009-01-01
Full Text Available In order to simulate neuronal electrical activities, we must estimate the dynamics of channel conductances from physiological experimental data. However, this approach requires the formulation of differential equations that express the time course of channel conductance. On the other hand, if the dynamics are automatically estimated, neuronal activities can be easily simulated. By using a recurrent neural network (RNN, it is possible to estimate the dynamics of channel conductances without formulating the differential equations. In the present study, we estimated the dynamics of the Na+ and K+ conductances of a squid giant axon using two different fully connected RNNs and were able to reproduce various neuronal activities of the axon. The reproduced activities were an action potential, a threshold, a refractory phenomenon, a rebound action potential, and periodic action potentials with a constant stimulation. RNNs can be trained using channels other than the Na+ and K+ channels. Therefore, using our RNN estimation method, the dynamics of channel conductance can be automatically estimated and the neuronal activities can be simulated using the channel RNNs. An RNN can be a useful tool to estimate the dynamics of the channel conductance of a neuron, and by using the method presented here, it is possible to simulate neuronal activities more easily than by using the previous methods.
A Neural-Dynamic Architecture for Concurrent Estimation of Object Pose and Identity
Directory of Open Access Journals (Sweden)
Oliver Lomp
2017-04-01
Full Text Available Handling objects or interacting with a human user about objects on a shared tabletop requires that objects be identified after learning from a small number of views and that object pose be estimated. We present a neurally inspired architecture that learns object instances by storing features extracted from a single view of each object. Input features are color and edge histograms from a localized area that is updated during processing. The system finds the best-matching view for the object in a novel input image while concurrently estimating the object’s pose, aligning the learned view with current input. The system is based on neural dynamics, computationally operating in real time, and can handle dynamic scenes directly off live video input. In a scenario with 30 everyday objects, the system achieves recognition rates of 87.2% from a single training view for each object, while also estimating pose quite precisely. We further demonstrate that the system can track moving objects, and that it can segment the visual array, selecting and recognizing one object while suppressing input from another known object in the immediate vicinity. Evaluation on the COIL-100 dataset, in which objects are depicted from different viewing angles, revealed recognition rates of 91.1% on the first 30 objects, each learned from four training views.
A Neural-Dynamic Architecture for Concurrent Estimation of Object Pose and Identity.
Lomp, Oliver; Faubel, Christian; Schöner, Gregor
2017-01-01
Handling objects or interacting with a human user about objects on a shared tabletop requires that objects be identified after learning from a small number of views and that object pose be estimated. We present a neurally inspired architecture that learns object instances by storing features extracted from a single view of each object. Input features are color and edge histograms from a localized area that is updated during processing. The system finds the best-matching view for the object in a novel input image while concurrently estimating the object's pose, aligning the learned view with current input. The system is based on neural dynamics, computationally operating in real time, and can handle dynamic scenes directly off live video input. In a scenario with 30 everyday objects, the system achieves recognition rates of 87.2% from a single training view for each object, while also estimating pose quite precisely. We further demonstrate that the system can track moving objects, and that it can segment the visual array, selecting and recognizing one object while suppressing input from another known object in the immediate vicinity. Evaluation on the COIL-100 dataset, in which objects are depicted from different viewing angles, revealed recognition rates of 91.1% on the first 30 objects, each learned from four training views.
Energy Technology Data Exchange (ETDEWEB)
Rosales-Colunga, Luis Manuel; De Leon Rodriguez, Antonio [Division de Biologia Molecular, Instituto Potosino de Investigacion Cientifica y Tecnologica, Camino a la Presa San Jose 2055, Col. Lomas 4a secc, San Luis Potosi, SLP 78216 (Mexico); Garcia, Raul Gonzalez [Centro de Investigacion y Estudios de Posgrado, Facultad de Ciencias Quimicas, Universidad Autonoma de San Luis Potosi, Av. Dr. Manuel Nava 6, San Luis Potosi, SLP 78210 (Mexico)
2010-12-15
Biological hydrogen production is an active research area due to the importance of this gas as an energy carrier and the advantages of using biological systems to produce it. A cheap and practical on-line hydrogen determination is desired in those processes. In this study, an artificial neural network (ANN) was developed to estimate the hydrogen production in fermentative processes. A back propagation neural network (BPNN) of one hidden layer with 12 nodes was selected. The BPNN training was done using the conjugated gradient algorithm and on-line measurements of dissolved CO{sub 2}, pH and oxidation-reduction potential during the fermentations of cheese whey by Escherichia coli {delta}hycA {delta}lacI (WDHL) strain with or without pH control. The correlation coefficient between the hydrogen production determined by gas chromatography and the hydrogen production estimated by the BPNN was 0.955. Results showed that the BPNN successfully estimated the hydrogen production using only on-line parameters in genetically modified E. coli fermentations either with or without pH control. This approach could be used for other hydrogen production systems. (author)
DEFF Research Database (Denmark)
Wang, Ting; Guan, Sheng-Uei; Puthusserypady, Sadasivan
2014-01-01
in the corresponding incremental way. Based on Single Discriminability (SD), where only the feature discrimination ability is computed, a new filter statistical feature discrimination ability predictive metric, called the Accumulative Discriminability (AD), is designed for the dynamical feature discrimination ability...... estimation. Moreover, a criterion that summarizes all the produced values of AD is employed with a GA (Genetic Algorithm)-based approach to obtain the optimum feature ordering for classification problems based on neural networks by means of IAL. Compared with the feature ordering obtained by other approaches...
A method to estimate emission rates from industrial stacks based on neural networks.
Olcese, Luis E; Toselli, Beatriz M
2004-11-01
This paper presents a technique based on artificial neural networks (ANN) to estimate pollutant rates of emission from industrial stacks, on the basis of pollutant concentrations measured on the ground. The ANN is trained on data generated by the ISCST3 model, widely accepted for evaluation of dispersion of primary pollutants as a part of an environmental impact study. Simulations using theoretical values and comparison with field data are done, obtaining good results in both cases at predicting emission rates. The application of this technique would allow the local environment authority to control emissions from industrial plants without need of performing direct measurements inside the plant. copyright 2004 Elsevier Ltd.
APPLYING ARTIFICIAL NEURAL NETWORK OPTIMIZED BY FIREWORKS ALGORITHM FOR STOCK PRICE ESTIMATION
Directory of Open Access Journals (Sweden)
Khuat Thanh Tung
2016-04-01
Full Text Available Stock prediction is to determine the future value of a company stock dealt on an exchange. It plays a crucial role to raise the profit gained by firms and investors. Over the past few years, many methods have been developed in which plenty of efforts focus on the machine learning framework achieving the promising results. In this paper, an approach based on Artificial Neural Network (ANN optimized by Fireworks algorithm and data preprocessing by Haar Wavelet is applied to estimate the stock prices. The system was trained and tested with real data of various companies collected from Yahoo Finance. The obtained results are encouraging.
Neural classifier in the estimation process of maturity of selected varieties of apples
Boniecki, P.; Piekarska-Boniecka, H.; Koszela, K.; Zaborowicz, M.; Przybył, K.; Wojcieszak, D.; Zbytek, Z.; Ludwiczak, A.; Przybylak, A.; Lewicki, A.
2015-07-01
This paper seeks to present methods of neural image analysis aimed at estimating the maturity state of selected varieties of apples which are popular in Poland. An identification of the degree of maturity of selected varieties of apples has been conducted on the basis of information encoded in graphical form, presented in the digital photos. The above process involves the application of the BBCH scale, used to determine the maturity of apples. The aforementioned scale is widely used in the EU and has been developed for many species of monocotyledonous plants and dicotyledonous plants. It is also worth noticing that the given scale enables detailed determinations of development stage of a given plant. The purpose of this work is to identify maturity level of selected varieties of apples, which is supported by the use of image analysis methods and classification techniques represented by artificial neural networks. The analysis of graphical representative features based on image analysis method enabled the assessment of the maturity of apples. For the utilitarian purpose the "JabVis 1.1" neural IT system was created, in accordance with requirements of the software engineering dedicated to support the decision-making processes occurring in broadly understood production process and processing of apples.
Porosity Estimation By Artificial Neural Networks Inversion . Application to Algerian South Field
Eladj, Said; Aliouane, Leila; Ouadfeul, Sid-Ali
2017-04-01
One of the main geophysicist's current challenge is the discovery and the study of stratigraphic traps, this last is a difficult task and requires a very fine analysis of the seismic data. The seismic data inversion allows obtaining lithological and stratigraphic information for the reservoir characterization . However, when solving the inverse problem we encounter difficult problems such as: Non-existence and non-uniqueness of the solution add to this the instability of the processing algorithm. Therefore, uncertainties in the data and the non-linearity of the relationship between the data and the parameters must be taken seriously. In this case, the artificial intelligence techniques such as Artificial Neural Networks(ANN) is used to resolve this ambiguity, this can be done by integrating different physical properties data which requires a supervised learning methods. In this work, we invert the acoustic impedance 3D seismic cube using the colored inversion method, then, the introduction of the acoustic impedance volume resulting from the first step as an input of based model inversion method allows to calculate the Porosity volume using the Multilayer Perceptron Artificial Neural Network. Application to an Algerian South hydrocarbon field clearly demonstrate the power of the proposed processing technique to predict the porosity for seismic data, obtained results can be used for reserves estimation, permeability prediction, recovery factor and reservoir monitoring. Keywords: Artificial Neural Networks, inversion, non-uniqueness , nonlinear, 3D porosity volume, reservoir characterization .
Directory of Open Access Journals (Sweden)
J. B. Habarulema
2012-05-01
Full Text Available In this work, results obtained by investigating the application of different neural network backpropagation training algorithms are presented. This was done to assess the performance accuracy of each training algorithm in total electron content (TEC estimations using identical datasets in models development and verification processes. Investigated training algorithms are standard backpropagation (SBP, backpropagation with weight delay (BPWD, backpropagation with momentum (BPM term, backpropagation with chunkwise weight update (BPC and backpropagation for batch (BPB training. These five algorithms are inbuilt functions within the Stuttgart Neural Network Simulator (SNNS and the main objective was to find out the training algorithm that generates the minimum error between the TEC derived from Global Positioning System (GPS observations and the modelled TEC data. Another investigated algorithm is the MatLab based Levenberg-Marquardt backpropagation (L-MBP, which achieves convergence after the least number of iterations during training. In this paper, neural network (NN models were developed using hourly TEC data (for 8 years: 2000–2007 derived from GPS observations over a receiver station located at Sutherland (SUTH (32.38° S, 20.81° E, South Africa. Verification of the NN models for all algorithms considered was performed on both "seen" and "unseen" data. Hourly TEC values over SUTH for 2003 formed the "seen" dataset. The "unseen" dataset consisted of hourly TEC data for 2002 and 2008 over Cape Town (CPTN (33.95° S, 18.47° E and SUTH, respectively. The models' verification showed that all algorithms investigated provide comparable results statistically, but differ significantly in terms of time required to achieve convergence during input-output data training/learning. This paper therefore provides a guide to neural network users for choosing appropriate algorithms based on the availability of computation capabilities used for research.
Directory of Open Access Journals (Sweden)
J. B. Habarulema
2012-05-01
Full Text Available In this work, results obtained by investigating the application of different neural network backpropagation training algorithms are presented. This was done to assess the performance accuracy of each training algorithm in total electron content (TEC estimations using identical datasets in models development and verification processes. Investigated training algorithms are standard backpropagation (SBP, backpropagation with weight delay (BPWD, backpropagation with momentum (BPM term, backpropagation with chunkwise weight update (BPC and backpropagation for batch (BPB training. These five algorithms are inbuilt functions within the Stuttgart Neural Network Simulator (SNNS and the main objective was to find out the training algorithm that generates the minimum error between the TEC derived from Global Positioning System (GPS observations and the modelled TEC data. Another investigated algorithm is the MatLab based Levenberg-Marquardt backpropagation (L-MBP, which achieves convergence after the least number of iterations during training. In this paper, neural network (NN models were developed using hourly TEC data (for 8 years: 2000–2007 derived from GPS observations over a receiver station located at Sutherland (SUTH (32.38° S, 20.81° E, South Africa. Verification of the NN models for all algorithms considered was performed on both "seen" and "unseen" data. Hourly TEC values over SUTH for 2003 formed the "seen" dataset. The "unseen" dataset consisted of hourly TEC data for 2002 and 2008 over Cape Town (CPTN (33.95° S, 18.47° E and SUTH, respectively. The models' verification showed that all algorithms investigated provide comparable results statistically, but differ significantly in terms of time required to achieve convergence during input-output data training/learning. This paper therefore provides a guide to neural network users for choosing appropriate algorithms based on the availability of computation capabilities used for research.
Besaw, L. E.; Rizzo, D. M.; Boitnoitt, G. N.
2006-12-01
Accurate, yet cost effective, sites characterization and analysis of uncertainty are the first steps in remediation efforts at sites with subsurface contamination. From the time of source identification to the monitoring and assessment of a remediation design, the management objectives change, resulting in increased costs and the need for additional data acquisition. Parameter estimation is a key component in reliable site characterization, contaminant flow and transport predictions, plume delineation and many other data management goals. We implement a data-driven parameter estimation technique using a counterpropagation Artificial Neural Network (ANN) that is able to incorporate multiple types of data. This method is applied to estimates of geophysical properties measured on a slab of Berea sandstone and delineation of the leachate plume migrating from a landfill in upstate N.Y. The estimates generated by the ANN have been found to be statistically similar to estimates generated using conventional geostatistical kriging methods. The associated parameter uncertainty in site characterization, due to sparsely distributed samples (spatial or temporal) and incomplete site knowledge, is of major concern in resource mining and environmental engineering. We also illustrate the ability of the ANN method to perform conditional simulation using the spatial structure of parameters identified with semi-variogram analysis. This method allows for the generation of simulations that respect the observed measurement data, as well as the data's underlying spatial structure. The method of conditional simulation is used in a 3-dimensional application to estimate the uncertainty of soil lithology.
Artificial Neural Networks and Gene Expression Programing based age estimation using facial features
Directory of Open Access Journals (Sweden)
Baddrud Z. Laskar
2015-10-01
Full Text Available This work is about estimating human age automatically through analysis of facial images. It has got a lot of real-world applications. Due to prompt advances in the fields of machine vision, facial image processing, and computer graphics, automatic age estimation via faces in computer is one of the dominant topics these days. This is due to widespread real-world applications, in areas of biometrics, security, surveillance, control, forensic art, entertainment, online customer management and support, along with cosmetology. As it is difficult to estimate the exact age, this system is to estimate a certain range of ages. Four sets of classifications have been used to differentiate a person’s data into one of the different age groups. The uniqueness about this study is the usage of two technologies i.e., Artificial Neural Networks (ANN and Gene Expression Programing (GEP to estimate the age and then compare the results. New methodologies like Gene Expression Programing (GEP have been explored here and significant results were found. The dataset has been developed to provide more efficient results by superior preprocessing methods. This proposed approach has been developed, tested and trained using both the methods. A public data set was used to test the system, FG-NET. The quality of the proposed system for age estimation using facial features is shown by broad experiments on the available database of FG-NET.
Directory of Open Access Journals (Sweden)
Gholam Reza Sheykhzadeh
2017-02-01
Full Text Available Introduction: Penetration resistance is one of the criteria for evaluating soil compaction. It correlates with several soil properties such as vehicle trafficability, resistance to root penetration, seedling emergence, and soil compaction by farm machinery. Direct measurement of penetration resistance is time consuming and difficult because of high temporal and spatial variability. Therefore, many different regressions and artificial neural network pedotransfer functions have been proposed to estimate penetration resistance from readily available soil variables such as particle size distribution, bulk density (Db and gravimetric water content (θm. The lands of Ardabil Province are one of the main production regions of potato in Iran, thus, obtaining the soil penetration resistance in these regions help with the management of potato production. The objective of this research was to derive pedotransfer functions by using regression and artificial neural network to predict penetration resistance from some soil variations in the agricultural soils of Ardabil plain and to compare the performance of artificial neural network with regression models. Materials and methods: Disturbed and undisturbed soil samples (n= 105 were systematically taken from 0-10 cm soil depth with nearly 3000 m distance in the agricultural lands of the Ardabil plain ((lat 38°15' to 38°40' N, long 48°16' to 48°61' E. The contents of sand, silt and clay (hydrometer method, CaCO3 (titration method, bulk density (cylinder method, particle density (Dp (pychnometer method, organic carbon (wet oxidation method, total porosity(calculating from Db and Dp, saturated (θs and field soil water (θf using the gravimetric method were measured in the laboratory. Mean geometric diameter (dg and standard deviation (σg of soil particles were computed using the percentages of sand, silt and clay. Penetration resistance was measured in situ using cone penetrometer (analog model at 10
An Artificial Neural Network for Movement Pattern Analysis to Estimate Blood Alcohol Content Level.
Gharani, Pedram; Suffoletto, Brian; Chung, Tammy; Karimi, Hassan A
2017-12-13
Impairments in gait occur after alcohol consumption, and, if detected in real-time, could guide the delivery of "just-in-time" injury prevention interventions. We aimed to identify the salient features of gait that could be used for estimating blood alcohol content (BAC) level in a typical drinking environment. We recruited 10 young adults with a history of heavy drinking to test our research app. During four consecutive Fridays and Saturdays, every hour from 8 p.m. to 12 a.m., they were prompted to use the app to report alcohol consumption and complete a 5-step straight-line walking task, during which 3-axis acceleration and angular velocity data was sampled at a frequency of 100 Hz. BAC for each subject was calculated. From sensor signals, 24 features were calculated using a sliding window technique, including energy, mean, and standard deviation. Using an artificial neural network (ANN), we performed regression analysis to define a model determining association between gait features and BACs. Part (70%) of the data was then used as a training dataset, and the results tested and validated using the rest of the samples. We evaluated different training algorithms for the neural network and the result showed that a Bayesian regularization neural network (BRNN) was the most efficient and accurate. Analyses support the use of the tandem gait task paired with our approach to reliably estimate BAC based on gait features. Results from this work could be useful in designing effective prevention interventions to reduce risky behaviors during periods of alcohol consumption.
Wireless Indoor Location Estimation Based on Neural Network RSS Signature Recognition (LENSR)
Energy Technology Data Exchange (ETDEWEB)
Kurt Derr; Milos Manic
2008-06-01
Location Based Services (LBS), context aware applications, and people and object tracking depend on the ability to locate mobile devices, also known as localization, in the wireless landscape. Localization enables a diverse set of applications that include, but are not limited to, vehicle guidance in an industrial environment, security monitoring, self-guided tours, personalized communications services, resource tracking, mobile commerce services, guiding emergency workers during fire emergencies, habitat monitoring, environmental surveillance, and receiving alerts. This paper presents a new neural network approach (LENSR) based on a competitive topological Counter Propagation Network (CPN) with k-nearest neighborhood vector mapping, for indoor location estimation based on received signal strength. The advantage of this approach is both speed and accuracy. The tested accuracy of the algorithm was 90.6% within 1 meter and 96.4% within 1.5 meters. Several approaches for location estimation using WLAN technology were reviewed for comparison of results.
A fuzzy neural network to estimate at completion costs of construction projects
Directory of Open Access Journals (Sweden)
Morteza Bagherpour
2012-04-01
Full Text Available In construction cost management system, normally earned value management (EVM is applied as an efficient control approach in both status detection and estimation at completion (EAC cost forecasting. The traditional approaches in EAC predictions normally extend the current situation of a project to the future by employing pervious performance factor. The proposed approach of this paper considers both qualitative and quantitative factors affecting the EAC prediction. The proposed approach of this research not only estimates the completion of the project, but also it can generate accurate forecast for the entire future periods using a fuzzy neural network model. The model is also implemented for a real-world case study and yields encouraging preliminary results.
Bayesian estimation inherent in a Mexican-hat-type neural network
Takiyama, Ken
2016-05-01
Brain functions, such as perception, motor control and learning, and decision making, have been explained based on a Bayesian framework, i.e., to decrease the effects of noise inherent in the human nervous system or external environment, our brain integrates sensory and a priori information in a Bayesian optimal manner. However, it remains unclear how Bayesian computations are implemented in the brain. Herein, I address this issue by analyzing a Mexican-hat-type neural network, which was used as a model of the visual cortex, motor cortex, and prefrontal cortex. I analytically demonstrate that the dynamics of an order parameter in the model corresponds exactly to a variational inference of a linear Gaussian state-space model, a Bayesian estimation, when the strength of recurrent synaptic connectivity is appropriately stronger than that of an external stimulus, a plausible condition in the brain. This exact correspondence can reveal the relationship between the parameters in the Bayesian estimation and those in the neural network, providing insight for understanding brain functions.
Improvement of radiation dose estimation due to nuclear accidents using deep neural network and GPU
Energy Technology Data Exchange (ETDEWEB)
Desterro, Filipe S.M.; Almeida, Adino A.H.; Pereira, Claudio M.N.A., E-mail: filipesantana18@gmail.com, E-mail: adino@ien.gov.br, E-mail: cmcoelho@ien.gov.br [Instituto de Engenharia Nuclear (IEN/CNEN-RJ), Rio de Janeiro, RJ (Brazil)
2017-07-01
Recently, the use of mobile devices has been proposed for dose assessment during nuclear accidents. The idea is to support field teams, providing an approximated estimation of the dose distribution map in the vicinity of the nuclear power plant (NPP), without needing to be connected to the NPP systems. In order to provide such stand-alone execution, the use of artificial neural networks (ANN) has been proposed in substitution of the complex and time consuming physical models executed by the atmospheric dispersion radionuclide (ADR) system. One limitation observed on such approach is the very time-consuming training of the ANNs. Moreover, if the number of input parameters increases the performance of standard ANNs, like Multilayer-Perceptron (MLP) with backpropagation training, is affected leading to unreasonable training time. To improve learning, allowing better dose estimations, more complex ANN architectures are required. ANNs with many layers (much more than a typical number of layers), referred to as Deep Neural Networks (DNN), for example, have demonstrating to achieve better results. On the other hand, the training of such ANNs is very much slow. In order to allow the use of such DNNs in a reasonable training time, a parallel programming solution, using Graphic Processing Units (GPU) and Computing Unified Device Architecture (CUDA) is proposed. This work focuses on the study of computational technologies for improvement of the ANNs to be used in the mobile application, as well as their training algorithms. (author)
Bayesian estimation inherent in a Mexican-hat-type neural network.
Takiyama, Ken
2016-05-01
Brain functions, such as perception, motor control and learning, and decision making, have been explained based on a Bayesian framework, i.e., to decrease the effects of noise inherent in the human nervous system or external environment, our brain integrates sensory and a priori information in a Bayesian optimal manner. However, it remains unclear how Bayesian computations are implemented in the brain. Herein, I address this issue by analyzing a Mexican-hat-type neural network, which was used as a model of the visual cortex, motor cortex, and prefrontal cortex. I analytically demonstrate that the dynamics of an order parameter in the model corresponds exactly to a variational inference of a linear Gaussian state-space model, a Bayesian estimation, when the strength of recurrent synaptic connectivity is appropriately stronger than that of an external stimulus, a plausible condition in the brain. This exact correspondence can reveal the relationship between the parameters in the Bayesian estimation and those in the neural network, providing insight for understanding brain functions.
Eppenhof, Koen A. J.; Pluim, Josien P. W.
2017-02-01
Error estimation in medical image registration is valuable when validating, comparing, or combining registration methods. To validate a nonlinear image registration method, ideally the registration error should be known for the entire image domain. We propose a supervised method for the estimation of a registration error map for nonlinear image registration. The method is based on a convolutional neural network that estimates the norm of the residual deformation from patches around each pixel in two registered images. This norm is interpreted as the registration error, and is defined for every pixel in the image domain. The network is trained using a set of artificially deformed images. Each training example is a pair of images: the original image, and a random deformation of that image. No manually labeled ground truth error is required. At test time, only the two registered images are required as input. We train and validate the network on registrations in a set of 2D digital subtraction angiography sequences, such that errors up to eight pixels can be estimated. We show that for this range of errors the convolutional network is able to learn the registration error in pairs of 2D registered images at subpixel precision. Finally, we present a proof of principle for the extension to 3D registration problems in chest CTs, showing that the method has the potential to estimate errors in 3D registration problems.
Directory of Open Access Journals (Sweden)
Pan-Sang Kang
2016-06-01
Full Text Available Polymer flooding is now considered a technically- and commercially-proven method for enhanced oil recovery (EOR. The viscosity of the injected polymer solution is the key property for successful polymer flooding. Given that the viscosity of a polymer solution has a non-linear relationship with various influential parameters (molecular weight, degree of hydrolysis, polymer concentration, cation concentration of polymer solution, shear rate, temperature and that measurement of viscosity based on these parameters is a time-consuming process, the range of solution samples and the measurement conditions need to be limited and precise. Viscosity estimation of the polymer solution is effective for these purposes. An artificial neural network (ANN was applied to the viscosity estimation of FlopaamTM 3330S, FlopaamTM 3630S and AN-125 solutions, three commonly-used EOR polymers. The viscosities measured and estimated by ANN and the Carreau model using Lee’s correlation, the only method for estimating the viscosity of an EOR polymer solution in unmeasured conditions, were compared. Estimation accuracy was evaluated by the average absolute relative deviation, which has been widely used for accuracy evaluation of the results of ANN models. In all conditions, the accuracy of the ANN model is higher than that of the Carreau model using Lee’s correlation.
Snow glacier melt estimation in tropical Andean glaciers using artificial neural networks
Directory of Open Access Journals (Sweden)
V. Moya Quiroga
2013-04-01
Full Text Available Snow and glacier melt (SGM estimation plays an important role in water resources management. Although melting process can be modelled by energy balance methods, such studies require detailed data, which is rarely available. Hence, new and simpler approaches are needed for SGM estimations. The present study aims at developing an artificial neural networks (ANN based technique for estimating the energy available for melt (EAM and SGM rates using available and easy to obtain data such as temperature, short-wave radiation and relative humidity. Several ANN and multiple linear regression models (MLR were developed to represent the energy fluxes and estimate the EAM. The models were trained using measured data from the Zongo glacier located in the outer tropics and validated against measured data from the Antizana glacier located in the inner tropics. It was found that ANN models provide a better generalisation when applied to other data sets. The performance of the models was improved by including Antizana data into the training set, as it was proved to provide better results than other techniques like the use of a prior logarithmic transformation. The final model was validated against measured data from the Alpine glaciers Argentière and Saint-Sorlin. Then, the models were applied for the estimation of SGM at Condoriri glacier. The estimated SGM was compared with SGM estimated by an enhanced temperature method and proved to have the same behaviour considering temperature sensibility. Moreover, the ANN models have the advantage of direct application, while the temperature method requires calibration of empirical coefficients.
Song, Dong; Wang, Haonan; Tu, Catherine Y.; Marmarelis, Vasilis Z.; Hampson, Robert E.; Deadwyler, Sam A.; Berger, Theodore W.
2013-01-01
One key problem in computational neuroscience and neural engineering is the identification and modeling of functional connectivity in the brain using spike train data. To reduce model complexity, alleviate overfitting, and thus facilitate model interpretation, sparse representation and estimation of functional connectivity is needed. Sparsities include global sparsity, which captures the sparse connectivities between neurons, and local sparsity, which reflects the active temporal ranges of the input-output dynamical interactions. In this paper, we formulate a generalized functional additive model (GFAM) and develop the associated penalized likelihood estimation methods for such a modeling problem. A GFAM consists of a set of basis functions convolving the input signals, and a link function generating the firing probability of the output neuron from the summation of the convolutions weighted by the sought model coefficients. Model sparsities are achieved by using various penalized likelihood estimations and basis functions. Specifically, we introduce two variations of the GFAM using a global basis (e.g., Laguerre basis) and group LASSO estimation, and a local basis (e.g., B-spline basis) and group bridge estimation, respectively. We further develop an optimization method based on quadratic approximation of the likelihood function for the estimation of these models. Simulation and experimental results show that both group-LASSO-Laguerre and group-bridge-B-spline can capture faithfully the global sparsities, while the latter can replicate accurately and simultaneously both global and local sparsities. The sparse models outperform the full models estimated with the standard maximum likelihood method in out-of-sample predictions. PMID:23674048
A Comparative Approach to Hand Force Estimation using Artificial Neural Networks
Directory of Open Access Journals (Sweden)
Farid Mobasser
2012-01-01
Full Text Available In many applications that include direct human involvement such as control of prosthetic arms, athletic training, and studying muscle physiology, hand force is needed for control, modeling and monitoring purposes. The use of inexpensive and easily portable active electromyography (EMG electrodes and position sensors would be advantageous in these applications compared to the use of force sensors which are often very expensive and require bulky frames. Among non-model-based estimation methods, Multilayer Perceptron Artificial Neural Networks (MLPANN has widely been used to estimate muscle force or joint torque from different anatomical features in humans or animals. This paper investigates the use of Radial Basis Function (RBF ANN and MLPANN for force estimation and experimentally compares the performance of the two methodologies for the same human anatomy, ie, hand force estimation, under an ensemble of operational conditions. In this unified study, the EMG signal readings from upper-arm muscles involved in elbow joint movement and elbow angular position and velocity are utilized as inputs to the ANNs. In addition, the use of the elbow angular acceleration signal as an input for the ANNs is also investigated.
Martínez-Martínez, Víctor; Baladrón, Carlos; Gomez-Gil, Jaime; Ruiz-Ruiz, Gonzalo; Navas-Gracia, Luis M; Aguiar, Javier M; Carro, Belén
2012-10-17
This paper presents a system based on an Artificial Neural Network (ANN) for estimating and predicting environmental variables related to tobacco drying processes. This system has been validated with temperature and relative humidity data obtained from a real tobacco dryer with a Wireless Sensor Network (WSN). A fitting ANN was used to estimate temperature and relative humidity in different locations inside the tobacco dryer and to predict them with different time horizons. An error under 2% can be achieved when estimating temperature as a function of temperature and relative humidity in other locations. Moreover, an error around 1.5 times lower than that obtained with an interpolation method can be achieved when predicting the temperature inside the tobacco mass as a function of its present and past values with time horizons over 150 minutes. These results show that the tobacco drying process can be improved taking into account the predicted future value of the monitored variables and the estimated actual value of other variables using a fitting ANN as proposed.
A Comparative Approach to Hand Force Estimation using Artificial Neural Networks.
Mobasser, Farid; Hashtrudi-Zaad, Keyvan
2012-01-01
In many applications that include direct human involvement such as control of prosthetic arms, athletic training, and studying muscle physiology, hand force is needed for control, modeling and monitoring purposes. The use of inexpensive and easily portable active electromyography (EMG) electrodes and position sensors would be advantageous in these applications compared to the use of force sensors which are often very expensive and require bulky frames. Among non-model-based estimation methods, Multilayer Perceptron Artificial Neural Networks (MLPANN) has widely been used to estimate muscle force or joint torque from different anatomical features in humans or animals. This paper investigates the use of Radial Basis Function (RBF) ANN and MLPANN for force estimation and experimentally compares the performance of the two methodologies for the same human anatomy, ie, hand force estimation, under an ensemble of operational conditions. In this unified study, the EMG signal readings from upper-arm muscles involved in elbow joint movement and elbow angular position and velocity are utilized as inputs to the ANNs. In addition, the use of the elbow angular acceleration signal as an input for the ANNs is also investigated.
A neural network model for estimating soil phosphorus using terrain analysis
Directory of Open Access Journals (Sweden)
Ali Keshavarzi
2015-12-01
Full Text Available Artificial neural network (ANN model was developed and tested for estimating soil phosphorus (P in Kouhin watershed area (1000 ha, Qazvin province, Iran using terrain analysis. Based on the soil distribution correlation, vegetation growth pattern across the topographically heterogeneous landscape, the topographic and vegetation attributes were used in addition to pedologic information for the development of ANN model in area for estimating of soil phosphorus. Totally, 85 samples were collected and tested for phosphorus contents and corresponding attributes were estimated by the digital elevation model (DEM. In order to develop the pedo-transfer functions, data linearity was checked, correlated and 80% was used for modeling and ANN was tested using 20% of collected data. Results indicate that 68% of the variation in soil phosphorus could be explained by elevation and Band 1 data and significant correlation was observed between input variables and phosphorus contents. There was a significant correlation between soil P and terrain attributes which can be used to derive the pedo-transfer function for soil P estimation to manage nutrient deficiency. Results showed that P values can be calculated more accurately with the ANN-based pedo-transfer function with the input topographic variables along with the Band 1.
Online estimation of Allan variance coefficients based on a neural-extended Kalman filter.
Miao, Zhiyong; Shen, Feng; Xu, Dingjie; He, Kunpeng; Tian, Chunmiao
2015-01-23
As a noise analysis method for inertial sensors, the traditional Allan variance method requires the storage of a large amount of data and manual analysis for an Allan variance graph. Although the existing online estimation methods avoid the storage of data and the painful procedure of drawing slope lines for estimation, they require complex transformations and even cause errors during the modeling of dynamic Allan variance. To solve these problems, first, a new state-space model that directly models the stochastic errors to obtain a nonlinear state-space model was established for inertial sensors. Then, a neural-extended Kalman filter algorithm was used to estimate the Allan variance coefficients. The real noises of an ADIS16405 IMU and fiber optic gyro-sensors were analyzed by the proposed method and traditional methods. The experimental results show that the proposed method is more suitable to estimate the Allan variance coefficients than the traditional methods. Moreover, the proposed method effectively avoids the storage of data and can be easily implemented using an online processor.
Visual Servoing from Deep Neural Networks
Bateux, Quentin; Marchand, Eric; Leitner, Jürgen; Chaumette, Francois; Corke, Peter
2017-01-01
International audience; We present a deep neural network-based method to perform high-precision, robust and real-time 6 DOF visual servoing. The paper describes how to create a dataset simulating various perturbations (occlusions and lighting conditions) from a single real-world image of the scene. A convolutional neural network is fine-tuned using this dataset to estimate the relative pose between two images of the same scene. The output of the network is then employed in a visual servoing c...
National Aeronautics and Space Administration — The GPM Ground Validation Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks Cloud Classification System (PERSIANN-CCS)...
Yarar, Alpaslan
2017-10-01
In this study, analytical and Artificial Neural Network (ANN) model were used for determine the discharge coefficient of Ogee Spillways. For this aim, discharge coefficients of 11 different heads were calculated by using a test flume having 7.5 cm width, 15 cm depth and 5 m length, in the laboratory. Discharge coefficients were also computed by the formula for the same heads measured in the laboratory to investigate the accuracy of experimental setup. An ANN model was set by using the experimental results in order to estimate the discharge coefficient. Then, the performance of the ANN model was investigated. As the result, the coefficient of determination between ANN model and experimental setup is found R2= 0.98. ANN model is show a good consistency with experimental results.
Directory of Open Access Journals (Sweden)
K Sedhuraman
2012-12-01
Full Text Available In this paper, a novel reactive power based model reference neural learning adaptive system (RP-MRNLAS is proposed. The model reference adaptive system (MRAS based speed estimation is one of the most popular methods used for sensor-less controlled induction motor drives. In conventional MRAS, the error adaptation is done using a Proportional-integral-(PI. The non-linear mapping capability of a neural network (NN and the powerful learning algorithms have increased the applications of NN in power electronics and drives. Thus, a neural learning algorithm is used for the adaptation mechanism in MRAS and is often referred to as a model reference neural learning adaptive system (MRNLAS. In MRNLAS, the error between the reference and neural learning adaptive models is back propagated to adjust the weights of the neural network for rotor speed estimation. The two different methods of MRNLAS are flux based (RF-MRNLAS and reactive power based (RP-MRNLAS. The reactive power- based methods are simple and free from integral equations as compared to flux based methods. The advantage of the reactive power based method and the NN learning algorithms are exploited in this work to yield a RPMRNLAS. The performance of the proposed RP-MRNLAS is analyzed extensively. The proposed RP-MRNLAS is compared in terms of accuracy and integrator drift problems with popular rotor flux-based MRNLAS for the same system and validated through Matlab/Simulink. The superiority of the RP- MRNLAS technique is demonstrated
Artificial neural network model for estimating the soil respiration under different land uses
Ebrahimi, Mitra; Sarikhani, Mohammad Reza; Safari Sinegani, Ali Akbar; Ahmadi, Abbas; Keesstra, Saskia
2017-04-01
Soil respiration is a biological process in microbes that convert organic carbon to atmospheric CO2. This process is considered to be one of the largest global carbon fluxes and is affected by different physicochemical and biological properties of soil, land usageuse, vegetation types and climate patterns. The aim of this study was to estimate the soil basal (BR) and substrate induced respiration (SIR) of 150 data obtained from soil samples collected from depth (0-25 cm) under different land uses by Artificial Neural Network. Soil samples were chosen from three provinces of Iran, with humid subtropical and semi-arid climate patterns. In each soil sample, soil texture, pH, electrical conductivity (EC), calcium carbonate equivalent (CCE), organic carbon (OC), OC fractionation data e.g. light fraction OC (LOC), heavy fraction OC (HOC), cold water extractable OC (COC) and warm water extractable OC (WOC), population of fungi, bacteria and actinomycete, BR and SIR were measured. Our goal was to use the most efficient ANN-model to predict soil respiration with simple soil data. Our results indicated that in an ANN model containing all the measured parameters, the R2 and RMSE values for BR prediction were 0.64 and 0.047 while these statistical indicators for SIR obtained 0.58 and 0.15, respectively. The R2 and RMSE values of the BR-ANN and SIR-ANN predicted models comprising 7 variables (including OC, pH, EC, CCE and soil texture) were estimated at 0.66, 0.043 and 0.52, 0.16, respectively. It was concluded that ANN modeling is a reliable method for predicting soil respiration. KEYWORDS: Artificial neural network; Land use; Soil physicochemical properties; Soil respiration; Soil microorganism
Modeling and adaptive control of a camless engine using neural networks and estimation techniques
Energy Technology Data Exchange (ETDEWEB)
Ashhab, S. [Hashemite Univ., Zarqa (Jordan). Dept. of Mechanical Engineering
2007-08-09
A system to control the cylinder air charge (CAC) in a camless internal combustion (IC) engine was recently developed. The performance of an IC engine connected to an adaptive artificial neural network (ANN) based feedback controller was then investigated. A control oriented model for the engine intake process was created based on thermodynamics laws and was validated against engine experimental data. Input-output data at a speed of 1500 RPM was generated and used to train an ANN model for the engine. The inputs were the intake valve lift (IVL) and closing timing (IVC). The output was the CAC. The controller consisted of a feedforward controller, CAC estimator, and on-line ANN parameter estimator. The feedforward controller provided IVL and IVC that satisfied the driver's torque demand and was the inverse of the engine ANN model. The on-line ANN used the error between the CAC measurement from the CAC estimator and its predicted value from the ANN to update the network's parameters. The feedforward controller was therefore adapted since its operation depended on the ANN model. The adaptation scheme improved the ANN prediction accuracy when the engine parts degraded, the speed changed or when modeling errors occurred. The engine controller exhibited good CAC tracking performance. Computer simulation demonstrated the capability of the camless engine controller. 17 refs., 5 figs.
[Estimation of Winter Wheat Biomass Using Visible Spectral and BP Based Artificial Neural Networks].
Cui, Ri-xian; Liu, Ya-dong; Fu, Jin-dong
2015-09-01
The objective of this study was to evaluate the feasibility of using color digital image analysis and back propagation (BP) based artificial neural networks (ANN) method to estimate above ground biomass at the canopy level of winter wheat field. Digital color images of winter wheat canopies grown under six levels of nitrogen treatments were taken with a digital camera for four times during the elongation stage and at the same time wheat plants were sampled to measure above ground biomass. Canopy cover (CC) and 10 color indices were calculated from winter wheat canopy images by using image analysis program (developed in Microsoft Visual Basic). Correlation analysis was carried out to identify the relationship between CC, 10 color indices and winter wheat above ground biomass. Stepwise multiple linear regression and BP based ANN methods were used to establish the models to estimate winter wheat above ground biomass. The results showed that CC, and two color indices had a significant cor- relation with above ground biomass. CC revealed the highest correlation with winter wheat above ground biomass. Stepwise multiple linear regression model constituting CC and color indices of NDI and b, and BP based ANN model with four variables (CC, g, b and NDI) for input was constructed to estimate winter wheat above ground biomass. The validation results indicate that the model using BP based ANN method has a better performance with higher R2 (0.903) and lower RMSE (61.706) and RRMSE (18.876) in comparation with the stepwise regression model.
Di Nuovo, Alessandro G; Di Nuovo, Santo; Buono, Serafino
2012-02-01
The estimation of a person's intelligence quotient (IQ) by means of psychometric tests is indispensable in the application of psychological assessment to several fields. When complex tests as the Wechsler scales, which are the most commonly used and universally recognized parameter for the diagnosis of degrees of retardation, are not applicable, it is necessary to use other psycho-diagnostic tools more suited for the subject's specific condition. But to ensure a homogeneous diagnosis it is necessary to reach a common metric, thus, the aim of our work is to build models able to estimate accurately and reliably the Wechsler IQ, starting from different psycho-diagnostic tools. Four different psychometric tests (Leiter international performance scale; coloured progressive matrices test; the mental development scale; psycho educational profile), along with the Wechsler scale, were administered to a group of 40 mentally retarded subjects, with various pathologies, and control persons. The obtained database is used to evaluate Wechsler IQ estimation models starting from the scores obtained in the other tests. Five modelling methods, two statistical and three from machine learning, that belong to the family of artificial neural networks (ANNs) are employed to build the estimator. Several error metrics for estimated IQ and for retardation level classification are defined to compare the performance of the various models with univariate and multivariate analyses. Eight empirical studies show that, after ten-fold cross-validation, best average estimation error is of 3.37 IQ points and mental retardation level classification error of 7.5%. Furthermore our experiments prove the superior performance of ANN methods over statistical regression ones, because in all cases considered ANN models show the lowest estimation error (from 0.12 to 0.9 IQ points) and the lowest classification error (from 2.5% to 10%). Since the estimation performance is better than the confidence interval of
Estimation of seismic quality factor: Artificial neural networks and current approaches
Yıldırım, Eray; Saatçılar, Ruhi; Ergintav, Semih
2017-01-01
The aims of this study are to estimate soil attenuation using alternatives to traditional methods, to compare results of using these methods, and to examine soil properties using the estimated results. The performances of all methods, amplitude decay, spectral ratio, Wiener filter, and artificial neural network (ANN) methods, are examined on field and synthetic data with noise and without noise. High-resolution seismic reflection field data from Yeniköy (Arnavutköy, İstanbul) was used as field data, and 424 estimations of Q values were made for each method (1,696 total). While statistical tests on synthetic and field data are quite close to the Q value estimation results of ANN, Wiener filter, and spectral ratio methods, the amplitude decay methods showed a higher estimation error. According to previous geological and geophysical studies in this area, the soil is water-saturated, quite weak, consisting of clay and sandy units, and, because of current and past landslides in the study area and its vicinity, researchers reported heterogeneity in the soil. Under the same physical conditions, Q value calculated on field data can be expected to be 7.9 and 13.6. ANN models with various structures, training algorithm, input, and number of neurons are investigated. A total of 480 ANN models were generated consisting of 60 models for noise-free synthetic data, 360 models for different noise content synthetic data and 60 models to apply to the data collected in the field. The models were tested to determine the most appropriate structure and training algorithm. In the final ANN, the input vectors consisted of the difference of the width, energy, and distance of seismic traces, and the output was Q value. Success rate of both ANN methods with noise-free and noisy synthetic data were higher than the other three methods. Also according to the statistical tests on estimated Q value from field data, the method showed results that are more suitable. The Q value can be estimated
Sai Gowtam, V.; Tulasi Ram, S.
2017-11-01
Artificial Neural Networks (ANNs) are known to be capable of solving linear as well as highly nonlinear problems. Using the long-term and high-quality data set of Formosa Satellite-3/Constellation Observing System for Meteorology, Ionosphere, and Climate (FORMOSAT-3/COSMIC, in short F3/C) from 2006 to 2015, an ANN-based two-dimensional (2-D) Ionospheric Model (ANNIM) is developed to predict the ionospheric peak parameters, such as NmF2 and hmF2. In this pilot study, the ANNIM results are compared with the original F3/C data, GRACE (Gravity Recovery and Climate Experiment) observations as well as International Reference Ionosphere (IRI)-2016 model to assess the learning efficiency of the neural networks used in the model. The ANNIM could well predict the NmF2 (hmF2) values with RMS errors of 1.87 × 105 el/cm3 (27.9 km) with respect to actual F3/C; and 2.98 × 105 el/cm3 (40.18 km) with respect to independent GRACE data. Further, the ANNIM predictions found to be as good as IRI-2016 model with a slightly smaller RMS error when compared to independent GRACE data. The ANNIM has successfully reproduced the local time, latitude, longitude, and seasonal variations with errors ranging 15-25% for NmF2 and 10-15% for hmF2 compared to actual F3/C data, except the postsunset enhancement in hmF2. Further, the ANNIM has also captured the global-scale ionospheric phenomena such as ionospheric annual anomaly, Weddell Sea Anomaly, and the midlatitude summer nighttime anomaly. Compared to IRI-2016 model, the ANNIM is found to have better represented the fine longitudinal structures and the midlatitude summer nighttime enhancements in both the hemispheres.
A hybrid neural networks-fuzzy logic-genetic algorithm for grade estimation.
Tahmasebi, Pejman; Hezarkhani, Ardeshir
2012-05-01
The grade estimation is a quite important and money/time-consuming stage in a mine project, which is considered as a challenge for the geologists and mining engineers due to the structural complexities in mineral ore deposits. To overcome this problem, several artificial intelligence techniques such as Artificial Neural Networks (ANN) and Fuzzy Logic (FL) have recently been employed with various architectures and properties. However, due to the constraints of both methods, they yield the desired results only under the specific circumstances. As an example, one major problem in FL is the difficulty of constructing the membership functions (MFs).Other problems such as architecture and local minima could also be located in ANN designing. Therefore, a new methodology is presented in this paper for grade estimation. This method which is based on ANN and FL is called "Coactive Neuro-Fuzzy Inference System" (CANFIS) which combines two approaches, ANN and FL. The combination of these two artificial intelligence approaches is achieved via the verbal and numerical power of intelligent systems. To improve the performance of this system, a Genetic Algorithm (GA) - as a well-known technique to solve the complex optimization problems - is also employed to optimize the network parameters including learning rate, momentum of the network and the number of MFs for each input. A comparison of these techniques (ANN, Adaptive Neuro-Fuzzy Inference System or ANFIS) with this new method (CANFIS-GA) is also carried out through a case study in Sungun copper deposit, located in East-Azerbaijan, Iran. The results show that CANFIS-GA could be a faster and more accurate alternative to the existing time-consuming methodologies for ore grade estimation and that is, therefore, suggested to be applied for grade estimation in similar problems.
Soft Sensors for Kerosene Properties Estimation and Control in Crude Distillation Unit
N. Bolf; Galinec, G.; Ivandić, M.
2009-01-01
Neural network-based soft sensors are developed for kerosene properties estimation, a refinery crude distillation unit side product. Based on temperature and flow measurements, two soft sensors serve as the estimators for the kerosene distillation end point (95 %) and freezing point. Soft sensor models are developed using linear regression techniques and neural networks. After performing multiple linear regression analysis it is determined that it is not possible to realize linear model...
Wang, Gang; Briskot, Till; Hahn, Tobias; Baumann, Pascal; Hubbuch, Jürgen
2017-03-03
Mechanistic modeling has been repeatedly successfully applied in process development and control of protein chromatography. For each combination of adsorbate and adsorbent, the mechanistic models have to be calibrated. Some of the model parameters, such as system characteristics, can be determined reliably by applying well-established experimental methods, whereas others cannot be measured directly. In common practice of protein chromatography modeling, these parameters are identified by applying time-consuming methods such as frontal analysis combined with gradient experiments, curve-fitting, or combined Yamamoto approach. For new components in the chromatographic system, these traditional calibration approaches require to be conducted repeatedly. In the presented work, a novel method for the calibration of mechanistic models based on artificial neural network (ANN) modeling was applied. An in silico screening of possible model parameter combinations was performed to generate learning material for the ANN model. Once the ANN model was trained to recognize chromatograms and to respond with the corresponding model parameter set, it was used to calibrate the mechanistic model from measured chromatograms. The ANN model's capability of parameter estimation was tested by predicting gradient elution chromatograms. The time-consuming model parameter estimation process itself could be reduced down to milliseconds. The functionality of the method was successfully demonstrated in a study with the calibration of the transport-dispersive model (TDM) and the stoichiometric displacement model (SDM) for a protein mixture. Copyright © 2017 The Author(s). Published by Elsevier B.V. All rights reserved.
Noise-insensitive no-reference image blur estimation by convolutional neural networks
Wegner, D.; Koerber, M.; Repasi, E.
2017-05-01
A few image quality metrics for blur assessment have been presented in the last years. However, most of those metrics do not take image noise into account. Yet, image noise is an unavoidable part of the image forming process with digital cameras. Some thermal imagers show larger sensor noise and inhomogeneity compared to cameras operating in the visible range. Further, natural imagery might contain a combination of several degradations. Assessment of degraded images by observer trials is expensive and time consuming. A single robust quality metric might be derived by metrics highly responsive to single degradations and insensitive to others. Hence separate assessment of image blur and noise seems to be reasonable. In this paper we present a deep learning approach for noise-insensitive blur predictions by using Convolutional Neural Networks (CNN) on image patches. In contrast to current blur metrics the model output is highly correlated to blur distortion over a wide range of image noise. The model is trained on images of ImageNet database impaired by Gaussian blur and noise and tested on artificial and natural image data. Local blur estimation based on patches is especially useful for estimation of non-uniform blur due to motion and atmospheric turbulence.
Energy Technology Data Exchange (ETDEWEB)
Garcia, Silvia R; Romo, Miguel P; Mayoral, Juan M [Instituto de Ingenieria, Universidad Nacional Autonoma de Mexico, Mexico D.F. (Mexico)
2007-01-15
An extensive analysis of the strong ground motion Mexican data base was conducted using Soft Computing (SC) techniques. A Neural Network NN is used to estimate both orthogonal components of the horizontal (PGAh) and vertical (PGAv) peak ground accelerations measured at rock sites during Mexican subduction zone earthquakes. The work discusses the development, training, and testing of this neural model. Attenuation phenomenon was characterized in terms of magnitude, epicentral distance and focal depth. Neural approximators were used instead of traditional regression techniques due to their flexibility to deal with uncertainty and noise. NN predictions follow closely measured responses exhibiting forecasting capabilities better than those of most established attenuation relations for the Mexican subduction zone. Assessment of the NN, was also applied to subduction zones in Japan and North America. For the database used in this paper the NN and the-better-fitted- regression approach residuals are compared. [Spanish] Un analisis exhaustivo de la base de datos mexicana de sismos fuertes se llevo a cabo utilizando tecnicas de computo aproximado, SC (soft computing). En particular, una red neuronal, NN, es utilizada para estimar ambos componentes ortogonales de la maxima aceleracion horizontal del terreno, PGAh, y la vertical, PGAv, medidas en sitios en roca durante terremotos generados en la zona de subduccion de la Republica Mexicana. El trabajo discute el desarrollo, entrenamiento, y prueba de este modelo neuronal. El fenomeno de atenuacion fue caracterizado en terminos de la magnitud, la distancia epicentral y la profundidad focal. Aproximaciones neuronales fueron utilizadas en lugar de tecnicas de regresion tradicionales por su flexibilidad para tratar con incertidumbre y ruido en los datos. La NN sigue de cerca la respuesta medida exhibiendo capacidades predictivas mejores que las mostradas por muchas de las relaciones de atenuacion establecidas para la zona de
Directory of Open Access Journals (Sweden)
Pouria Behnoud far
2017-09-01
Full Text Available Lost circulation can cause an increase in time and cost of operation. Pipe sticking, formation damage and uncontrolled flow of oil and gas may be consequences of lost circulation. Dealing with this problem is a key factor to conduct a successful drilling operation. Estimation of lost circulation amount is necessary to find a solution. Lost circulation is influenced by different parameters such as mud weight, pump pressure, depth etc. Mud weight, pump pressure and flow rate of mud should be designed to prevent induced fractures and have the least amount of lost circulation. Artificial neural network is useful to find the relations of parameters with lost circulation. Genetic algorithm is applied on the achieved relations to determine the optimum mud weight, pump pressure, and flow rate. In an Iranian oil field, daily drilling reports of wells which are drilled using UBD technique are studied. Asmari formation is the most important oil reservoir of the studied field and UBD is used only in this interval. Three wells with the most, moderate and without lost circulation are chosen. In this article, the effect of mud weight, depth, pump pressure and flow rate of pump on lost circulation in UBD of Asmari formation in one of the Southwest Iranian fields is studied using drilling data and artificial neural network. In addition, the amount of lost circulation is predicted precisely with respect to two of the studied parameters using the presented correlations and the optimum mud weight, pump pressure and flow rate are calculated to minimize the lost circulation amount.
A. R. Moghadassi; Parvizian,F.; Hosseini, S.M.; Fazlali,A. R.
2009-01-01
Equations of state are useful for description of fluid properties such as pressure-volume-temperature (PVT). However, the success estimation of such correlations depends mainly on the range of data which have originated. Therefore new models are highly required. In this work a new method is proposed based on Artificial Neural Network (ANN) for estimation of PVT properties of compounds. The data sets were collected from Perry's Chemical Engineers' Handbook. Different training schemes for the b...
Directory of Open Access Journals (Sweden)
Wen-Yeau Chang
2013-01-01
Full Text Available This paper proposes an equivalent circuit parameters measurement and estimation method for proton exchange membrane fuel cell (PEMFC. The parameters measurement method is based on current loading technique; in current loading test a no load PEMFC is suddenly turned on to obtain the waveform of the transient terminal voltage. After the equivalent circuit parameters were measured, a hybrid method that combines a radial basis function (RBF neural network and enhanced particle swarm optimization (EPSO algorithm is further employed for the equivalent circuit parameters estimation. The RBF neural network is adopted such that the estimation problem can be effectively processed when the considered data have different features and ranges. In the hybrid method, EPSO algorithm is used to tune the connection weights, the centers, and the widths of RBF neural network. Together with the current loading technique, the proposed hybrid estimation method can effectively estimate the equivalent circuit parameters of PEMFC. To verify the proposed approach, experiments were conducted to demonstrate the equivalent circuit parameters estimation of PEMFC. A practical PEMFC stack was purposely created to produce the common current loading activities of PEMFC for the experiments. The practical results of the proposed method were studied in accordance with the conditions for different loading conditions.
Directory of Open Access Journals (Sweden)
Zhong Min Su
2011-04-01
Full Text Available We propose a generalized regression neural network (GRNN approach based on grey relational analysis (GRA and principal component analysis (PCA (GP-GRNN to improve the accuracy of density functional theory (DFT calculation for homolysis bond dissociation energies (BDE of Y-NO bond. As a demonstration, this combined quantum chemistry calculation with the GP-GRNN approach has been applied to evaluate the homolysis BDE of 92 Y-NO organic molecules. The results show that the full-descriptor GRNN without GRA and PCA (F-GRNN and with GRA (G-GRNN approaches reduce the root-mean-square (RMS of the calculated homolysis BDE of 92 organic molecules from 5.31 to 0.49 and 0.39 kcal mol−1 for the B3LYP/6-31G (d calculation. Then the newly developed GP-GRNN approach further reduces the RMS to 0.31 kcal mol−1. Thus, the GP-GRNN correction on top of B3LYP/6-31G (d can improve the accuracy of calculating the homolysis BDE in quantum chemistry and can predict homolysis BDE which cannot be obtained experimentally.
Estimates of global and regional prevalence of neural tube defects for 2015: a systematic analysis.
Blencowe, Hannah; Kancherla, Vijaya; Moorthie, Sowmiya; Darlison, Matthew W; Modell, Bernadette
2018-01-24
Neural tube defects (NTDs) are associated with substantial mortality, morbidity, disability, and psychological and economic costs. Many are preventable with folic acid, and access to appropriate services for those affected can improve survival and quality of life. We used a compartmental model to estimate global and regional birth prevalence of NTDs (live births, stillbirths, and elective terminations of pregnancy) and subsequent under-5 mortality. Data were identified through web-based reviews of birth defect registry databases and systematic literature reviews. Meta-analyses were undertaken where appropriate. For 2015, our model estimated 260,100 (uncertainty interval (UI): 213,800-322,000) NTD-affected birth outcomes worldwide (prevalence 18.6 (15.3-23.0)/10,000 live births). Approximately 50% of cases were elective terminations of pregnancy for fetal anomalies (UI: 59,300 (47,900-74,500)) or stillbirths (57,800 (UI: 35,000-88,600)). Of NTD-affected live births, 117,900 (∼75%) (UI: 105,500-186,600) resulted in under-5 deaths. Our systematic review showed a paucity of high-quality data in the regions of the world with the highest burden. Despite knowledge about prevention, NTDs remain highly prevalent worldwide. Lack of surveillance and incomplete ascertainment of affected pregnancies make NTDs invisible to policy makers. Improved surveillance of all adverse outcomes is needed to improve the robustness of total NTD prevalence estimation, evaluate effectiveness of prevention through folic acid fortification, and improve outcomes through care and rehabilitation. © 2018 The Authors. Annals of the New York Academy of Sciences published by Wiley Periodicals Inc. on behalf of The New York Academy of Sciences.
Directory of Open Access Journals (Sweden)
Raphaëlle Sauzède
2017-05-01
Full Text Available A neural network-based method (CANYON: CArbonate system and Nutrients concentration from hYdrological properties and Oxygen using a Neural-network was developed to estimate water-column (i.e., from surface to 8,000 m depth biogeochemically relevant variables in the Global Ocean. These are the concentrations of three nutrients [nitrate (NO3−, phosphate (PO43−, and silicate (Si(OH4] and four carbonate system parameters [total alkalinity (AT, dissolved inorganic carbon (CT, pH (pHT, and partial pressure of CO2 (pCO2], which are estimated from concurrent in situ measurements of temperature, salinity, hydrostatic pressure, and oxygen (O2 together with sampling latitude, longitude, and date. Seven neural-networks were developed using the GLODAPv2 database, which is largely representative of the diversity of open-ocean conditions, hence making CANYON potentially applicable to most oceanic environments. For each variable, CANYON was trained using 80 % randomly chosen data from the whole database (after eight 10° × 10° zones removed providing an “independent data-set” for additional validation, the remaining 20 % data were used for the neural-network test of validation. Overall, CANYON retrieved the variables with high accuracies (RMSE: 1.04 μmol kg−1 (NO3−, 0.074 μmol kg−1 (PO43−, 3.2 μmol kg−1 (Si(OH4, 0.020 (pHT, 9 μmol kg−1 (AT, 11 μmol kg−1 (CT and 7.6 % (pCO2 (30 μatm at 400 μatm. This was confirmed for the eight independent zones not included in the training process. CANYON was also applied to the Hawaiian Time Series site to produce a 22 years long simulated time series for the above seven variables. Comparison of modeled and measured data was also very satisfactory (RMSE in the order of magnitude of RMSE from validation test. CANYON is thus a promising method to derive distributions of key biogeochemical variables. It could be used for a variety of global and regional applications ranging from data quality control
Directory of Open Access Journals (Sweden)
Eyad K Almaita
2017-03-01
Keywords: Energy efficiency, Power quality, Radial basis function, neural networks, adaptive, harmonic. Article History: Received Dec 15, 2016; Received in revised form Feb 2nd 2017; Accepted 13rd 2017; Available online How to Cite This Article: Almaita, E.K and Shawawreh J.Al (2017 Improving Stability and Convergence for Adaptive Radial Basis Function Neural Networks Algorithm (On-Line Harmonics Estimation Application. International Journal of Renewable Energy Develeopment, 6(1, 9-17. http://dx.doi.org/10.14710/ijred.6.1.9-17
Directory of Open Access Journals (Sweden)
Amir LAKZIAN
2010-09-01
Full Text Available This paper presents the comparison of three different approaches to estimate soil water content at defined values of soil water potential based on selected parameters of soil solid phase. Forty different sampling locations in northeast of Iran were selected and undisturbed samples were taken to measure the water content at field capacity (FC, -33 kPa, and permanent wilting point (PWP, -1500 kPa. At each location solid particle of each sample including the percentage of sand, silt and clay were measured. Organic carbon percentage and soil texture were also determined for each soil sample at each location. Three different techniques including pattern recognition approach (k nearest neighbour, k-NN, Artificial Neural Network (ANN and pedotransfer functions (PTF were used to predict the soil water at each sampling location. Mean square deviation (MSD and its components, index of agreement (d, root mean square difference (RMSD and normalized RMSD (RMSDr were used to evaluate the performance of all the three approaches. Our results showed that k-NN and PTF performed better than ANN in prediction of water content at both FC and PWP matric potential. Various statistics criteria for simulation performance also indicated that between kNN and PTF, the former, predicted water content at PWP more accurate than PTF, however both approach showed a similar accuracy to predict water content at FC.
Fuzzy neural network model for the estimation of subpixel land cover composition
Binaghi, Elisabetta; Brivio, Pietro A.; Ghezzi, Pier P.; Rampini, Anna; Vicenzi, Massimo
1998-12-01
This paper reports on an experimental study designed for the in-depth investigation of how a supervised neuro-fuzzy classifier evaluates partial membership in land cover classes. The system is based on the Fuzzy Multilayer Perceptron model proposed by Pal and Mitra to which modifications in distance measures adopted for computing gradual membership to fuzzy class are introduced. During the training phase supervised learning is used to assign output class membership to pure training vectors (full membership to one land cover class); the model supports a procedure to automatically compute fuzzy output membership values for mixed training pixels. The classifier has been evaluated by conducting two experiments. The first employed simulated tests images which include pure and mixed pixels of known geometry and radiometry. The second experiment was conducted on a highly complex real scene of the Venice lagoon, Italy) where water and wetland merge into one another, at sub-pixel level. Accuracy of the results produced by the classifier was evaluated and compared using evaluation tools specifically defined and implemented to extend conventional descriptive and analytical statistical estimators to the case of multi-membership in classes. Results obtained demonstrated in the specific context of mixed pixels that the classification benefits from the integration of neural and fuzzy techniques.
Broiler weight estimation based on machine vision and artificial neural network.
Amraei, S; Abdanan Mehdizadeh, S; Salari, S
2017-04-01
1. Machine vision and artificial neural network (ANN) procedures were used to estimate live body weight of broiler chickens in 30 1-d-old broiler chickens reared for 42 d. 2. Imaging was performed two times daily. To localise chickens within the pen, an ellipse fitting algorithm was used and the chickens' head and tail removed using the Chan-Vese method. 3. The correlations between the body weight and 6 physical extracted features indicated that there were strong correlations between body weight and the 5 features including area, perimeter, convex area, major and minor axis length. 5. According to statistical analysis there was no significant difference between morning and afternoon data over 42 d. 6. In an attempt to improve the accuracy of live weight approximation different ANN techniques, including Bayesian regulation, Levenberg-Marquardt, Scaled conjugate gradient and gradient descent were used. Bayesian regulation with R 2 value of 0.98 was the best network for prediction of broiler weight. 7. The accuracy of the machine vision technique was examined and most errors were less than 50 g.
Neural effects of social environmental stress - an activation likelihood estimation meta-analysis.
Mothersill, O; Donohoe, G
2016-07-01
Social environmental stress, including childhood abuse and deprivation, is associated with increased rates of psychiatric disorders such as schizophrenia and depression. However, the neural mechanisms mediating risk are not completely understood. Functional magnetic resonance imaging (MRI) studies have reported effects of social environmental stress on a variety of brain regions, but interpretation of results is complicated by the variety of environmental risk factors examined and different methods employed. We examined brain regions consistently showing differences in blood oxygen level-dependent (BOLD) response in individuals exposed to higher levels of environmental stress by performing a coordinate-based meta-analysis on 54 functional MRI studies using activation likelihood estimation (ALE), including an overall sample of 3044 participants. We performed separate ALE analyses on studies examining adults (mean age ⩾18 years) and children/adolescents (mean age environmental stress across multiple studies. These clusters incorporated several brain regions, among which the right amygdala was most frequently implicated. These findings suggest that a variety of social environmental stressors is associated with differences in the BOLD response of specific brain regions such as the right amygdala in both children/adolescents and adults. What remains unknown is whether these environmental stressors have differential effects on treatment response in these brain regions.
Sensorless speed estimation of an AC induction motor by using an artificial neural network approach
Alkhoraif, Abdulelah Ali
Sensorless speed detection of an induction motor is an attractive area for researchers to enhance the reliability of the system and to reduce the cost of the components. This paper presents a simple method of estimating a rotational speed by utilizing an artificial neural network (ANN) that would be fed by a set of stator current frequencies that contain some saliency harmonics. This approach allows operators to detect the speed in induction motors such an approach also provides reliability, low cost, and simplicity. First, the proposed method is based on converting the stator current signals to the frequency domain and then applying a tracking algorithm to the stator current spectrum in order to detect frequency peaks. Secondly, the ANN has to be trained by the detected peaks; the training data must be from very precise data to provide an accurate rotor speed. Moreover, the desired output of the training is the speed, which is measured by a tachometer simultaneously with the stator current signal. The databases were collected at many different speeds from two different types of AC induction motors, wound rotor and squirrel cage. They were trained and tested, so when the difference between the desired speed value and the ANN output value reached the wanted accuracy, the system does not need to use the tachometer anymore. Eventually, the experimental results show that in an optimal ANN design, the speed of the wound rotor induction motor was estimated accurately, where the testing average error was 1 RPM. The proposed method has not succeeded to predict the rotor speed of the squirrel cage induction motor precisely, where the smallest testingaverage error that was achieved was 5 RPM.
DEFF Research Database (Denmark)
Soliman, Hammam Abdelaal Hammam; Wang, Huai; Blaabjerg, Frede
2016-01-01
of the aforementioned challenges and shortcomings. In this paper, a pure software condition monitoring method based on Artificial Neural Network (ANN) algorithm is proposed. The implemented ANN estimates the capacitance of the dc-link capacitor in a back-to-back converter. The error analysis of the estimated results......The reliability of dc-link capacitors in power electronic converters is one of the critical aspects to be considered in modern power converter design. The observation of their ageing process and the estimation of their health status have been an attractive subject for the industrial field and hence...
Short-term estimation of GNSS TEC using a neural network model in Brazil
Ferreira, Arthur Amaral; Borges, Renato Alves; Paparini, Claudia; Ciraolo, Luigi; Radicella, Sandro M.
2017-10-01
This work presents a novel Neural Network (NN) model to estimate Total Electron Content (TEC) from Global Navigation Satellite Systems (GNSS) measurements in three distinct sectors in Brazil. The purpose of this work is to start the investigations on the development of a regional model that can be used to determine the vertical TEC over Brazil, aiming future applications on a near real-time frame estimations and short-term forecasting. The NN is used to estimate the GNSS TEC values at void locations, where no dual-frequency GNSS receiver that may be used as a source of data to GNSS TEC estimation is available. This approach is particularly useful for GNSS single-frequency users that rely on corrections of ionospheric range errors by TEC models. GNSS data from the first GLONASS network for research and development (GLONASS R&D network) installed in Latin America, and from the Brazilian Network for Continuous Monitoring of the GNSS (RMBC) were used on TEC calibration. The input parameters of the NN model are based on features known to influence TEC values, such as geographic location of the GNSS receiver, magnetic activity, seasonal and diurnal variations, and solar activity. Data from two ten-days periods (from DoY 154 to 163 and from 282 to 291) are used to train the network. Three distinct analyses have been carried out in order to assess time-varying and spatial performance of the model. At the spatial performance analysis, for each region, a set of stations is chosen to provide training data to the NN, and after the training procedure, the NN is used to estimate vTEC behavior for the test station which data were not presented to the NN in training process. An analysis is done by comparing, for each testing station, the estimated NN vTEC delivered by the NN and reference calibrated vTEC. Also, as a second analysis, the network ability to forecast one day after the time interval (DoY 292) based on information of the second period of investigation is also assessed
Directory of Open Access Journals (Sweden)
N. Baghdadi
2012-06-01
Full Text Available The purpose of this study was to develop an approach to estimate soil surface parameters from C-band polarimetric SAR data in the case of bare agricultural soils. An inversion technique based on multi-layer perceptron (MLP neural networks was introduced. The neural networks were trained and validated on a noisy simulated dataset generated from the Integral Equation Model (IEM on a wide range of surface roughness and soil moisture, as it is encountered in agricultural contexts for bare soils. The performances of neural networks in retrieving soil moisture and surface roughness were tested for several inversion cases using or not using a-priori knowledge on soil parameters. The inversion approach was then validated using RADARSAT-2 images in polarimetric mode. The introduction of expert knowledge on the soil moisture (dry to wet soils or very wet soils improves the soil moisture estimates, whereas the precision on the surface roughness estimation remains unchanged. Moreover, the use of polarimetric parameters α_{1} and anisotropy were used to improve the soil parameters estimates. These parameters provide to neural networks the probable ranges of soil moisture (lower or higher than 0.30 cm^{3} cm^{−3} and surface roughness (root mean square surface height lower or higher than 1.0 cm. Soil moisture can be retrieved correctly from C-band SAR data by using the neural networks technique. Soil moisture errors were estimated at about 0.098 cm^{3} cm^{−3} without a-priori information on soil parameters and 0.065 cm^{3} cm^{−3} (RMSE applying a-priori information on the soil moisture. The retrieval of surface roughness is possible only for low and medium values (lower than 2 cm. Results show that the precision on the soil roughness estimates was about 0.7 cm. For surface roughness lower than 2 cm, the precision on the soil roughness is better with an RMSE about 0.5 cm. The use of polarimetric
Neural and Neural Gray-Box Modeling for Entry Temperature Prediction in a Hot Strip Mill
Barrios, José Angel; Torres-Alvarado, Miguel; Cavazos, Alberto; Leduc, Luis
2011-10-01
In hot strip mills, initial controller set points have to be calculated before the steel bar enters the mill. Calculations rely on the good knowledge of rolling variables. Measurements are available only after the bar has entered the mill, and therefore they have to be estimated. Estimation of process variables, particularly that of temperature, is of crucial importance for the bar front section to fulfill quality requirements, and the same must be performed in the shortest possible time to preserve heat. Currently, temperature estimation is performed by physical modeling; however, it is highly affected by measurement uncertainties, variations in the incoming bar conditions, and final product changes. In order to overcome these problems, artificial intelligence techniques such as artificial neural networks and fuzzy logic have been proposed. In this article, neural network-based systems, including neural-based Gray-Box models, are applied to estimate scale breaker entry temperature, given its importance, and their performance is compared to that of the physical model used in plant. Several neural systems and several neural-based Gray-Box models are designed and tested with real data. Taking advantage of the flexibility of neural networks for input incorporation, several factors which are believed to have influence on the process are also tested. The systems proposed in this study were proven to have better performance indexes and hence better prediction capabilities than the physical models currently used in plant.
Directory of Open Access Journals (Sweden)
David Eriksson
2016-08-01
Full Text Available To test the importance of a certain cell type or brain area it is common to make a lack of function experiment in which the neuronal population of interest is inhibited. Here we review physiological and methodological constraints for making controlled perturbations using the corticothalamic circuit as an example. The brain with its many types of cells and rich interconnectivity offers many paths through which a perturbation can spread within a short time. To understand the side effects of the perturbation one should record from those paths. We find that ephaptic effects, gap-junctions, and fast chemical synapses are so fast that they can react to the perturbation during the few milliseconds it takes for an opsin to change the membrane potential. The slow chemical synapses, astrocytes, extracellular ions and vascular signals, will continue to give their physiological input for around 20 milliseconds before they also react to the perturbation. Although we show that some pathways can react within milliseconds the strength/speed reported in this review should be seen as an upper bound since we have omitted how polysynaptic signals are attenuated. Thus the number of additional recordings that has to be made to control for the perturbation side effects is expected to be fewer than proposed here. To summarize, the reviewed literature not only suggests that it is possible to make controlled lack of function experiments, but, it also suggests that such a lack of function experiment can be used to measure the context of local neural computations.
Estimation of Leak Flow Rate during Post-LOCA Using Cascaded Fuzzy Neural Networks
Energy Technology Data Exchange (ETDEWEB)
Kim, Dong Yeong [Korea Atomic Energy Research Institute, Daejeon (Korea, Republic of); Na, Man Gyun [Chosun University, Gwangju (Korea, Republic of)
2016-10-15
In this study, important parameters such as the break position, size, and leak flow rate of loss of coolant accidents (LOCAs), provide operators with essential information for recovering the cooling capability of the nuclear reactor core, for preventing the reactor core from melting down, and for managing severe accidents effectively. Leak flow rate should consist of break size, differential pressure, temperature, and so on (where differential pressure means difference between internal and external reactor vessel pressure). The leak flow rate is strongly dependent on the break size and the differential pressure, but the break size is not measured and the integrity of pressure sensors is not assured in severe circumstances. In this paper, a cascaded fuzzy neural network (CFNN) model is appropriately proposed to estimate the leak flow rate out of break, which has a direct impact on the important times (time approaching the core exit temperature that exceeds 1200 .deg. F, core uncover time, reactor vessel failure time, etc.). The CFNN is a data-based model, it requires data to develop and verify itself. Because few actual severe accident data exist, it is essential to obtain the data required in the proposed model using numerical simulations. In this study, a CFNN model was developed to predict the leak flow rate before proceeding to severe LOCAs. The simulations showed that the developed CFNN model accurately predicted the leak flow rate with less error than 0.5%. The CFNN model is much better than FNN model under the same conditions, such as the same fuzzy rules. At the result of comparison, the RMS errors of the CFNN model were reduced by approximately 82 ~ 97% of those of the FNN model.
Estimating sediment-fill thickness in intermontane valleys using artificial neural networks
Mey, Juergen; Scherler, Dirk; Strecker, Manfred R.; Zeilinger, Gerold
2014-05-01
Knowledge about the thickness and spatial distribution of sedimentary fills in intermontane valleys and internally drained orogenic plateau settings is important for many applications in the fields of hydrology, geology, geohazards, economic resources, and geomorphology. However, direct measurements of sediment thickness are time consuming and require sophisticated geophysical tools, infrastructure and logistics that are often not available. This has resulted in a general scarcity of such data, often incomplete and fraught with error, especially in remote areas. Here, we present a new approach to estimate valley fill thickness based on the geometric properties of a landscape using artificial neural networks. We test the potential of this approach by employing a 3-stage procedure. First, we run tests with three synthetic datasets representing valleys involving different complexities to explore our model's sensitivity to network architecture and training data. Second, we apply the method to a glacierized setting in the European Alps in the region of the Unteraar Glacier and the Rhone Glacier where ground-penetrating radar measurements of ice thickness allow for an analysis of the prediction performance in active subglacial terrain. In the third step, we estimate the sediment-fill thickness of the Rhone Valley, one of the largest intermontane basins in the Western Alps, where seismic reflection data are used as a benchmark for an assessment of the method's performance on a large spatial scale of a formerly glaciated landscape. Our results to date show that a successful application strongly depends on the network architecture and the choice of the training region. For the currently ice-covered catchments our prediction of the ice thickness is in good agreement with the geophysical control data, but below that of methods, which are based on ice-mechanical principles. However, in case of the sedimentary fill of the Rhone Valley where no corresponding physically
Using convolutional neural networks to estimate time-of-flight from PET detector waveforms
Berg, Eric; Cherry, Simon R.
2018-01-01
Although there have been impressive strides in detector development for time-of-flight positron emission tomography, most detectors still make use of simple signal processing methods to extract the time-of-flight information from the detector signals. In most cases, the timing pick-off for each waveform is computed using leading edge discrimination or constant fraction discrimination, as these were historically easily implemented with analog pulse processing electronics. However, now with the availability of fast waveform digitizers, there is opportunity to make use of more of the timing information contained in the coincident detector waveforms with advanced signal processing techniques. Here we describe the application of deep convolutional neural networks (CNNs), a type of machine learning, to estimate time-of-flight directly from the pair of digitized detector waveforms for a coincident event. One of the key features of this approach is the simplicity in obtaining ground-truth-labeled data needed to train the CNN: the true time-of-flight is determined from the difference in path length between the positron emission and each of the coincident detectors, which can be easily controlled experimentally. The experimental setup used here made use of two photomultiplier tube-based scintillation detectors, and a point source, stepped in 5 mm increments over a 15 cm range between the two detectors. The detector waveforms were digitized at 10 GS s‑1 using a bench-top oscilloscope. The results shown here demonstrate that CNN-based time-of-flight estimation improves timing resolution by 20% compared to leading edge discrimination (231 ps versus 185 ps), and 23% compared to constant fraction discrimination (242 ps versus 185 ps). By comparing several different CNN architectures, we also showed that CNN depth (number of convolutional and fully connected layers) had the largest impact on timing resolution, while the exact network parameters, such as convolutional
Directory of Open Access Journals (Sweden)
A. Babaei
2013-06-01
Full Text Available In this paper a method for estimating the dimension of rectangular cracks is proposed. The use of Eddy current (ECnondestructive testing (NDT based on probe impedance changes on the crack regions is considered. The artificialneural network estimates the dimension of new cracks using impedance changes of the eddy current probe. Theexperimental results and finite element method (FEM results are used for training the artificial neural network. Byincreasing the number of experiments, the results of the finite element method are not necessary. The simulationresults are very promising.
Energy Technology Data Exchange (ETDEWEB)
Almonacid, F.; Rus, C.; Perez-Higueras, P.; Hontoria, L.
2010-07-01
Despite the great technological advances in photovoltaic and in particular in network-connected systems, efforts are still required in research, technological development and innovation (i + d + i) must be aimed primarily at addressing the different system parts. one aspect that can help achieve this goal is majorette estimation methods of energy produced by photovoltaic generators. There are a number of cases resulting in a decrease of the expected energy. In this paper we will compare a standard method widely used in the estimation of the power of the photovoltaic generator with another novel method, developed at the University of Jaen, based on artificial neural networks (ANN). (Author) 9 refs.
Directory of Open Access Journals (Sweden)
A. R. Moghadassi
2009-03-01
Full Text Available Equations of state are useful for description of fluid properties such as pressure-volume-temperature (PVT. However, the success estimation of such correlations depends mainly on the range of data which have originated. Therefore new models are highly required. In this work a new method is proposed based on Artificial Neural Network (ANN for estimation of PVT properties of compounds. The data sets were collected from Perry's Chemical Engineers' Handbook. Different training schemes for the back-propagation learning algorithm, such as; Scaled Conjugate Gradient (SCG, Levenberg-Marquardt (LM and Resilient back Propagation (RP methods were used. The accuracy and trend stability of the trained networks were tested against unseen data. The LM algorithm with sixty neurons in the hidden layer has proved to be the best suitable algorithm with the minimum Mean Square Error (MSE of 0.000606. The ANN's capability to estimate the PVT properties is one of the best estimating method with high performance.
Directory of Open Access Journals (Sweden)
Dongoun Lee
2016-09-01
Full Text Available Due to the increasing costs of construction waste disposal, an accurate estimation of the amount of construction waste is a key factor in a project’s success. Korea has been burdened by increasing construction waste as a consequence of the growing number of construction projects and a lack of construction waste management (CWM strategies. One of the problems associated with predicting the amount of waste is that there are no suitable estimation strategies currently available. Therefore, we developed a hybrid estimation model to predict the quantity and cost of waste in the early stage of construction. The proposed approach can be used to address cost overruns and improve CWM in the subsequent stages of construction. The proposed hybrid model uses artificial neural networks (ANNs and ant colony optimization (ACO. It is expected to provide an accurate waste estimate by applying historical data from multifamily residential buildings.
Geraldo Ferreira, A.; Soria, Emilio; Lopez-Baeza, Ernesto; Vila, Joan; Serrano, Antonio J.; Martinez, Marcelino; Velazquez Blazquez, Almudena; Clerbaux, Nicolas
This paper describes the results obtained using Artificial Neural Networks (AAN) models to estimate the diurnal cycle of net radiation (Rn) at surface. The data used as input parameter in the AAN model were that measured by Geostationary Earth Radiation Budget (GERB-1) instrument, on board Meteosat 9 satellite. The data concerning Rn at the surface were collected at the Valencia Anchor Station (VAS), a ground reference meteorological station for the validation of low spatial resolution sensors situated near de city of Valencia, Spain. This data refers to the periods July 31st -August 6th 2006 and June 19th -August 18th 2007. Both, GERB-1 and VAS data are used to train and validate the AAN model. The same data set is also used to develop and validate a Multivariate Linear Regression (MLR) model. A comparison between the estimates provided by the AAN and the MLR models has been carried out; the results obtained with the neural model outperform the linear model. Moreover, the low values of the error indexes show that neural models can be used as an alternative methodology to make atmospheric corrections.
Directory of Open Access Journals (Sweden)
Ayedh Alqahtani
2013-09-01
Full Text Available Industrial application of life-cycle cost analysis (LCCA is somewhat limited, with techniques deemed overly theoretical, resulting in a reluctance to realise (and pass onto the client the advantages to be gained from objective (LCCA comparison of (subcomponent material specifications. To address the need for a user-friendly structured approach to facilitate complex processing, the work described here develops a new, accessible framework for LCCA of construction projects; it acknowledges Artificial Neural Networks (ANNs to compute the whole-cost(s of construction and uses the concept of cost significant items (CSI to identify the main cost factors affecting the accuracy of estimation. ANNs is a powerful means to handle non-linear problems and subsequently map between complex input/output data, address uncertainties. A case study documenting 20 building projects was used to test the framework and estimate total running costs accurately. Two methods were used to develop a neural network model; firstly a back-propagation method was adopted (using MATLAB SOFTWARE; and secondly, spread-sheet optimisation was conducted (using Microsoft Excel Solver. The best network was established as consisting of 19 hidden nodes, with the tangent sigmoid used as a transfer function of NNs model for both methods. The results find that in both neural network models, the accuracy of the developed NNs model is 1% (via Excel-solver and 2% (via back-propagation respectively.
Artificial neural networks based predictive model for worker ...
African Journals Online (AJOL)
Virtual cellular manufacturing systems (VCMS) have come into existence, replacing traditional cellular manufacturing systems (CMS), to meet highly dynamic production conditions in terms of demand, processing times, product mix and processing sequence. While cell formation phase of VCMS has been dealt quite ...
Prediction of epitopes using neural network based methods
DEFF Research Database (Denmark)
Lundegaard, Claus; Lund, Ole; Nielsen, Morten
2011-01-01
In this paper, we describe the methodologies behind three different aspects of the NetMHC family for prediction of MHC class I binding, mainly to HLAs. We have updated the prediction servers, NetMHC-3.2, NetMHCpan-2.2, and a new consensus method, NetMHCcons, which, in their previous versions, hav...
A Neural Network Based Workstation for Automated Cell Proliferation Analysis
2001-10-25
proliferation analysis, of cytological microscope images. The software of the system assists the expert biotechnologist during cell proliferation and...work was supported by the Programa de Apoyo a Proyectos de Desarrollo e Investigacíon en Informática REDII 2000. We thank Blanca Itzel Taboada for
Artificial neural network based modeling and controlling of ...
African Journals Online (AJOL)
Further for developing the model has been selected suitable adaptive filter. Two types of ... The simulation results revealed that NN-IMC with appropriate learning rate - momentum is capable to pursue the set-point changes and to reject the disturbance changes without steady state error or oscillations. NN-IMC with inverse ...
Neural network based system for script identification in Indian ...
Indian Academy of Sciences (India)
Author Affiliations. S Basavaraj Patil1 N V Subbareddy1 2. Kuvempu University Research Centre, Department of Computer Science and Engineering, University B D T College of Engineering, Davangere 577 004, India; Department of Computer Science & Engineering, Manipal Institute of Technology, Manipal 576 119, India ...
neural network based load frequency control for restructuring power
African Journals Online (AJOL)
2012-03-01
Mar 1, 2012 ... power system is chosen and load frequency con- trol of this system is made by a ANN controller and a conventional PI controller. Basically, power system consists of a governor, a turbine, and a generator with feedback of reg- ulation constant. System also includes step load change input to the generator.
A neural network based seafloor classification using acoustic backscatter
Digital Repository Service at National Institute of Oceanography (India)
Chakraborty, B.
stream_size 9 stream_content_type text/plain stream_name Adv_Soft_Comput_AFSS_2002_245.pdf.txt stream_source_info Adv_Soft_Comput_AFSS_2002_245.pdf.txt Content-Encoding ISO-8859-1 Content-Type text/plain; charset=ISO-8859-1 ...
Directory of Open Access Journals (Sweden)
Tosun Erdi
2017-01-01
Full Text Available This study was aimed at estimating the variation of several engine control parameters within the rotational speed-load map, using regression analysis and artificial neural network techniques. Duration of injection, specific fuel consumption, exhaust gas at turbine inlet, and within the catalytic converter brick were chosen as the output parameters for the models, while engine speed and brake mean effective pressure were selected as independent variables for prediction. Measurements were performed on a turbocharged direct injection spark ignition engine fueled with gasoline. A three-layer feed-forward structure and back-propagation algorithm was used for training the artificial neural network. It was concluded that this technique is capable of predicting engine parameters with better accuracy than linear and non-linear regression techniques.
Directory of Open Access Journals (Sweden)
CRUZ A. J. G.
1999-01-01
Full Text Available This paper reports on the use of the gas balance and dynamic methods to obtain an estimate of the volumetric oxygen transfer coefficient (kLa in a conventional reactor during the growth phase of the microorganism Cephalosporium acremonium. A new way of calculating kLa by the dynamic method employing an electrode with a slow response, is proposed. The calculated values of kLa were used in the training of a feedforward neural network, for which the inputs were the parameter measurements of the related variables. The neural network technique proved effective, predicting values of kLa accurately from input data not used during the training phase. In contrast, the gas balance method was shown to be less useful. This could be attributed to the poor data obtained with the apparatus used to measure the oxygen in the exhaust gas, explained by the low rate of oxygen consumption by the microorganism.
Building a Network Based Laboratory Environment
Directory of Open Access Journals (Sweden)
Sea Shuan Luo
2009-12-01
Full Text Available This paper presents a comparative study about the development of a network based laboratory environment in the “Unix introduction” course for the undergraduate students. The study results and the response from the students from 2005 to 2006 will be used to better understand what kind of method is more suitable for students. We also use the data collected to adjust our teaching strategy and try to build up a network based laboratory environment.
Estimating and suppressing background in Raman spectra with an artificial neural network
DEFF Research Database (Denmark)
Sigurdsson, Sigurdur; Larsen, Jan; Philipsen, Peter Alshede
2003-01-01
In this report we address the problem of skin fluorescence in feature extraction from Raman spectra of skin lesions. We apply a highly automated neural network method for suppressing skin fluorescence from Raman spectrum of skin lesions before dimension reduction with principal components analysis....... By applying the background suppression, the effect of outlier spectrum in the principal components analysis was greatly reduced....
Hotson, Guy; Smith, Ryan J; Rouse, Adam G; Schieber, Marc H; Thakor, Nitish V; Wester, Brock A
2016-07-01
Brain-machine interfaces (BMIs) are a rapidly progressing technology with the potential to restore function to victims of severe paralysis via neural control of robotic systems. Great strides have been made in directly mapping a user's cortical activity to control of the individual degrees of freedom of robotic end-effectors. While BMIs have yet to achieve the level of reliability desired for widespread clinical use, environmental sensors (e.g. RGB-D cameras for object detection) and prior knowledge of common movement trajectories hold great potential for improving system performance. Here we present a novel sensor fusion paradigm for BMIs that capitalizes on information able to be extracted from the environment to greatly improve the performance of control. This was accomplished by using dynamic movement primitives to model the 3D endpoint trajectories of manipulating various objects. We then used a switching unscented Kalman filter to continuously arbitrate between the 3D endpoint kinematics predicted by the dynamic movement primitives and control derived from neural signals. We experimentally validated our system by decoding 3D endpoint trajectories executed by a non-human primate manipulating four different objects at various locations. Performance using our system showed a dramatic improvement over using neural signals alone, with median distance between actual and decoded trajectories decreasing from 31.1 cm to 9.9 cm, and mean correlation increasing from 0.80 to 0.98. Our results indicate that our sensor fusion framework can dramatically increase the fidelity of neural prosthetic trajectory decoding.
Vargas-Meléndez, Leandro; Boada, Beatriz L; Boada, María Jesús L; Gauchía, Antonio; Díaz, Vicente
2016-08-31
This article presents a novel estimator based on sensor fusion, which combines the Neural Network (NN) with a Kalman filter in order to estimate the vehicle roll angle. The NN estimates a "pseudo-roll angle" through variables that are easily measured from Inertial Measurement Unit (IMU) sensors. An IMU is a device that is commonly used for vehicle motion detection, and its cost has decreased during recent years. The pseudo-roll angle is introduced in the Kalman filter in order to filter noise and minimize the variance of the norm and maximum errors' estimation. The NN has been trained for J-turn maneuvers, double lane change maneuvers and lane change maneuvers at different speeds and road friction coefficients. The proposed method takes into account the vehicle non-linearities, thus yielding good roll angle estimation. Finally, the proposed estimator has been compared with one that uses the suspension deflections to obtain the pseudo-roll angle. Experimental results show the effectiveness of the proposed NN and Kalman filter-based estimator.
DEFF Research Database (Denmark)
Dasgupta, Sakyasingha; Goldschmidt, Dennis; Wörgötter, Florentin
2015-01-01
movements, (2) distributed (at each leg) recurrent neural network based adaptive forward models with efference copies as internal models for sensory predictions and instantaneous state estimations, and (3) searching and elevation control for adapting the movement of an individual leg to deal with different...... conditions, like uneven terrains, gaps, obstacles etc. Biological study has revealed that such complex behaviors are a result of a combination of biomechanics and neural mechanism thus representing the true nature of embodied interactions. While the biomechanics helps maintain flexibility and sustain...... a variety of movements, the neural mechanisms generate movements while making appropriate predictions crucial for achieving adaptation. Such predictions or planning ahead can be achieved by way of internal models that are grounded in the overall behavior of the animal. Inspired by these findings, we present...
Estimating Of Etchant Copper Concentration In The Electrolytic Cell Using Artificial Neural Networks
Directory of Open Access Journals (Sweden)
Muzher M. Ibrahem
2015-02-01
Full Text Available In this paper, Artificial Neural Networks (ANN, which are known for their ability to model nonlinear systems, provide accurate approximations of system behavior and are typically much more computationally efficient than phenomenological models are used to predict the etchant copper concentration in the electrolytic cell in terms of electric potential, operating time, temperature of the electrolytic cell , ratio of surface area of poles per unit volume of solution and the distance between poles. In this paper 350 sets of data are used to trained and test the network.. The best results were achieved using a model based on a feedforword Artificial Neural Network (ANN with one hidden layer and fifteen neurons in the hidden layer gives a very close prediction of the copper concentration in the electrolytic cell.
Genetic algorithm pruning of probabilistic neural networks in medical disease estimation.
Mantzaris, Dimitrios; Anastassopoulos, George; Adamopoulos, Adam
2011-10-01
A hybrid model consisting of an Artificial Neural Network (ANN) and a Genetic Algorithm procedure for diagnostic risk factors selection in Medicine is proposed in this paper. A medical disease prediction may be viewed as a pattern classification problem based on a set of clinical and laboratory parameters. Probabilistic Neural Network models were assessed in terms of their classification accuracy concerning medical disease prediction. A Genetic Algorithm search was performed to examine potential redundancy in the diagnostic factors. This search led to a pruned ANN architecture, minimizing the number of diagnostic factors used during the training phase and therefore minimizing the number of nodes in the ANN input and hidden layer as well as the Mean Square Error of the trained ANN at the testing phase. As a conclusion, a number of diagnostic factors in a patient's data record can be omitted without loss of fidelity in the diagnosis procedure. Copyright © 2011 Elsevier Ltd. All rights reserved.
Non-Linear State Estimation Using Pre-Trained Neural Networks
DEFF Research Database (Denmark)
Bayramoglu, Enis; Andersen, Nils Axel; Ravn, Ole
2010-01-01
effecting the transformation. This function is approximated by a neural network using offline training. The training is based on monte carlo sampling. A way to obtain parametric distributions of flexible shape to be used easily with these networks is also presented. The method can also be used to improve...... other parametric methods around regions with strong non-linearities by including them inside the network....
A Computational Estimation of Cyclic Material Properties Using Artificial Neural Networks
Tomasella, A.; Dsoki, C. el; H. Hanselka; Kaufmann, H.
2011-01-01
The structural durability design of components requires the knowledge of cyclic material properties. These parameters are strongly dependent on environmental conditions and manufacturing processes, and require many experimental tests to be correctly determined. Considering time and costs, it is not possible to include in the tests all the variables that influence the material behaviour. For this reason, the computational method of the Artificial Neural Network (ANN) can be implemented to supp...
Network-based functional enrichment
Directory of Open Access Journals (Sweden)
Poirel Christopher L
2011-11-01
Full Text Available Abstract Background Many methods have been developed to infer and reason about molecular interaction networks. These approaches often yield networks with hundreds or thousands of nodes and up to an order of magnitude more edges. It is often desirable to summarize the biological information in such networks. A very common approach is to use gene function enrichment analysis for this task. A major drawback of this method is that it ignores information about the edges in the network being analyzed, i.e., it treats the network simply as a set of genes. In this paper, we introduce a novel method for functional enrichment that explicitly takes network interactions into account. Results Our approach naturally generalizes Fisher’s exact test, a gene set-based technique. Given a function of interest, we compute the subgraph of the network induced by genes annotated to this function. We use the sequence of sizes of the connected components of this sub-network to estimate its connectivity. We estimate the statistical significance of the connectivity empirically by a permutation test. We present three applications of our method: i determine which functions are enriched in a given network, ii given a network and an interesting sub-network of genes within that network, determine which functions are enriched in the sub-network, and iii given two networks, determine the functions for which the connectivity improves when we merge the second network into the first. Through these applications, we show that our approach is a natural alternative to network clustering algorithms. Conclusions We presented a novel approach to functional enrichment that takes into account the pairwise relationships among genes annotated by a particular function. Each of the three applications discovers highly relevant functions. We used our methods to study biological data from three different organisms. Our results demonstrate the wide applicability of our methods. Our algorithms are
Krinitskiy, Mikhail; Sinitsyn, Alexey
2017-04-01
Shortwave radiation is an important component of surface heat budget over sea and land. To estimate them accurate observations of cloud conditions are needed including total cloud cover, spatial and temporal cloud structure. While massively observed visually, for building accurate SW radiation parameterizations cloud structure needs also to be quantified using precise instrumental measurements. While there already exist several state of the art land-based cloud-cameras that satisfy researchers needs, their major disadvantages are associated with inaccuracy of all-sky images processing algorithms which typically result in the uncertainties of 2-4 octa of cloud cover estimates with the resulting true-scoring cloud cover accuracy of about 7%. Moreover, none of these algorithms determine cloud types. We developed an approach for cloud cover and structure estimating, which provides much more accurate estimates and also allows for measuring additional characteristics. This method is based on the synthetic controlling index, namely the "grayness rate index", that we introduced in 2014. Since then this index has already demonstrated high efficiency being used along with the technique namely the "background sunburn effect suppression", to detect thin clouds. This made it possible to significantly increase the accuracy of total cloud cover estimation in various sky image states using this extension of routine algorithm type. Errors for the cloud cover estimates significantly decreased down resulting the mean squared error of about 1.5 octa. Resulting true-scoring accuracy is more than 38%. The main source of this approach uncertainties is the solar disk state determination errors. While the deep neural networks approach lets us to estimate solar disk state with 94% accuracy, the final result of total cloud estimation still isn`t satisfying. To solve this problem completely we applied the set of machine learning algorithms to the problem of total cloud cover estimation
Samui, Saumyadip; Samui Pal, Shanoli
2017-02-01
We present an improved photometric redshift estimator code, CuBANz, that is publicly available at https://goo.gl/fpk90V. It uses the back propagation neural network along with clustering of the training set, which makes it more efficient than existing neural network codes. In CuBANz, the training set is divided into several self learning clusters with galaxies having similar photometric properties and spectroscopic redshifts within a given span. The clustering algorithm uses the color information (i.e. u - g , g - r etc.) rather than the apparent magnitudes at various photometric bands as the photometric redshift is more sensitive to the flux differences between different bands rather than the actual values. Separate neural networks are trained for each cluster using all possible colors, magnitudes and uncertainties in the measurements. For a galaxy with unknown redshift, we identify the closest possible clusters having similar photometric properties and use those clusters to get the photometric redshifts using the particular networks that were trained using those cluster members. For galaxies that do not match with any training cluster, the photometric redshifts are obtained from a separate network that uses entire training set. This clustering method enables us to determine the redshifts more accurately. SDSS Stripe 82 catalog has been used here for the demonstration of the code. For the clustered sources with redshift range zspec < 0.7, the residual error (〈(zspec -zphot) 2 〉 1 / 2) in the training/testing phase is as low as 0.03 compared to the existing ANNz code that provides residual error on the same test data set of 0.05. Further, we provide a much better estimate of the uncertainty of the derived photometric redshift.
Shekarrizfard, Maryam; Karimi-Jashni, A; Hadad, K
2012-01-01
In this paper, a novel method in the estimation and prediction of PM(10) is introduced using wavelet transform-based artificial neural networks (WT-ANN). First, the application of wavelet transform, selected for its temporal shift properties and multiresolution analysis characteristics enabling it to reduce disturbing perturbations in input training set data, is presented. Afterward, the circular statistical indices which are used in this method are formally introduced in order to investigate the relation between PM(10) levels and circular meteorological variables. Then, the results of the simulation of PM(10) based on WT-ANN by use of MATLAB software are discussed. The results of the above-mentioned simulation show an enhanced accuracy and speed in PM(10) estimation/prediction and a high degree of robustness compared with traditional ANN models.
Jing, Yuanshu; Ruthaikarn, Buaphean; Jin, Xinyi; Pang, Bo
The evapotranspiration estimation is a key item for irrigation program. It has the important practical significance for high stable yield and water-saving in the region of red soil, South China. Penman-Monteith equation, recommended by FAO, is verified to be the most effective calculation to actual evaporation in many regions of the world. The only default is it has to use complete meteorological factors. To solve this problem, we are trying to find out a artificial neural network model (ANN) which can easily get its information and easy to calculate as well as guaranteed accuracy. A Bowen ratio energy balance (BREB) system and automatic weather station were employed for simultaneous measurement of actual evapotranspiration above the rice field. The frequency of 20-min recording provided the possibility for the estimation of daily and hourly evapotranspiration. The determined coefficient from the artificial neural network model on daily scale R2 is 0.9642, while hourly scale R2 is 0.9880. The reason was that the hourly scale training samples was greater than the daily scale measures. In general, the model gives an effective and feasible way for the evaluation of paddy rice evapotranspiration by the conventional parameters.