Performance prediction for Grid workflow activities based on features-ranked RBF network
Wang Jie; Duan Rubing; Farrukh Nadeem
2009-01-01
Accurate performance prediction of Grid workflow activities can help Grid schedulers map activities to appropriate Grid sites. This paper describes an approach based on features-ranked RBF neural network to predict the performance of Grid workflow activities. Experimental results for two kinds of real world Grid workflow activities are presented to show effectiveness of our approach.
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%.
Improving the Input of Classified Neural Networks Through Feature Construction
无
2001-01-01
A general classification algorithm of neural networks is unable to obtain satisfied results because of the uncertain problems existing among the features in most classification programs, such as interaction. With new features constructed by optimizing decision trees of examples, the input of neural networks is improved and an optimized classification algorithm based on neural networks is presented. A concept of dispersion of a classification program is also introduced too in this paper. At the end of the paper, an analysis is made with an example.``
Research on Method of Character Recognition Based on Hough Transform and RBF Neural Network
Zhang Yin
2015-01-01
Full Text Available A method of character recognition based on Hough transform and RBF neural network is proposed through research on weight accumulation algorithm of Hough transform. According to the feature of characters’ structure by using the duality of point-line Hough transform was done. In this method, the number of the points on the same line in parameter space and the position coordinates of the elements in image mapping space were taken to RBF neural network recognition system as characteristic input vector. It reduced the dimension of character feature vector and reflected the overall distribution of character lattice and the essential feature of character shape. The simulation results indicated there were some merits in this improved method: capability of recognition is strong, the quantity of calculation is small, and the speed of calculation is quick.
RBF neural network and active circles based algorithm for contours extraction
Zhou Zhiheng; Zeng Delu; Xie Shengli
2007-01-01
For the contours extraction from the images, active contour model and self-organizing map based approach are popular nowadays. But they are still confronted with the problems that the optimization of energy function will trap in local minimums and the contour evolutions greatly depend on the initial contour selection. Addressing to these problems, a contours extraction algorithm based on RBF neural network is proposed here. A series of circles with adaptive radius and center is firstly used to search image feature points that are scattered enough. After the feature points are clustered, a group of radial basis functions are constructed. Using the pixels' intensities and gradients as the input vector, the final object contour can be obtained by the predicting ability of the neural network. The RBF neural network based algorithm is tested on three kinds of images, such as changing topology, complicated background, and blurring or noisy boundary. Simulation results show that the proposed algorithm performs contours extraction greatly.
Cortical network reorganization guided by sensory input features.
Kilgard, Michael P; Pandya, Pritesh K; Engineer, Navzer D; Moucha, Raluca
2002-12-01
Sensory experience alters the functional organization of cortical networks. Previous studies using behavioral training motivated by aversive or rewarding stimuli have demonstrated that cortical plasticity is specific to salient inputs in the sensory environment. Sensory experience associated with electrical activation of the basal forebrain (BasF) generates similar input specific plasticity. By directly engaging plasticity mechanisms and avoiding extensive behavioral training, BasF stimulation makes it possible to efficiently explore how specific sensory features contribute to cortical plasticity. This review summarizes our observations that cortical networks employ a variety of strategies to improve the representation of the sensory environment. Different combinations of receptive-field, temporal, and spectrotemporal plasticity were generated in primary auditory cortex neurons depending on the pitch, modulation rate, and order of sounds paired with BasF stimulation. Simple tones led to map expansion, while modulated tones altered the maximum cortical following rate. Exposure to complex acoustic sequences led to the development of combination-sensitive responses. This remodeling of cortical response characteristics may reflect changes in intrinsic cellular mechanisms, synaptic efficacy, and local neuronal connectivity. The intricate relationship between the pattern of sensory activation and cortical plasticity suggests that network-level rules alter the functional organization of the cortex to generate the most behaviorally useful representation of the sensory environment.
Liu, Lin; Shen, Songhua; Liu, Qiang
2006-11-01
A novel method to detect power quality disturbance of distribution power system combing complex wavelet transform (WT) with radial basis function (RBF) neural network is presented. The paper tries to explain to design complex supported orthogonal wavelets by Morlet compactly supported orthogonal real wavelets, and then explore the extraction of disturbance signal to obtain the feature information, and finally propose several novel wavelet combined information (CI) to analyze the disturbance signal, superior to real wavelet analysis result. The feature obtained from WT coefficients are inputted into RBF network for power quality disturbance pattern recognition. The power quality disturbance recognition model is established and the synthesized method of recursive orthogonal least squares algorithm (ROLSA) with improved Givens transform is used to fulfill the network structure and parameter identification. By means of choosing enough samples to train the recognition model, the type of disturbance can be obtained when signal representing fault is inputted to the trained network. The results of simulation analysis show that the complex WT combined with RBF network are more sensitive to signal singularity, and found to be significant improvement over current methods in real-time detection and better noise proof ability.
Wear Debris Identification Using Feature Extraction and Neural Network
王伟华; 马艳艳; 殷勇辉; 王成焘
2004-01-01
A method and results of identification of wear debris using their morphological features are presented. The color images of wear debris were used as initial data. Each particle was characterized by a set of numerical parameters combined by its shape, color and surface texture features through a computer vision system. Those features were used as input vector of artificial neural network for wear debris identification. A radius basis function (RBF) network based model suitable for wear debris recognition was established,and its algorithm was presented in detail. Compared with traditional recognition methods, the RBF network model is faster in convergence, and higher in accuracy.
Evolving RBF neural networks for adaptive soft-sensor design.
Alexandridis, Alex
2013-12-01
This work presents an adaptive framework for building soft-sensors based on radial basis function (RBF) neural network models. The adaptive fuzzy means algorithm is utilized in order to evolve an RBF network, which approximates the unknown system based on input-output data from it. The methodology gradually builds the RBF network model, based on two separate levels of adaptation: On the first level, the structure of the hidden layer is modified by adding or deleting RBF centers, while on the second level, the synaptic weights are adjusted with the recursive least squares with exponential forgetting algorithm. The proposed approach is tested on two different systems, namely a simulated nonlinear DC Motor and a real industrial reactor. The results show that the produced soft-sensors can be successfully applied to model the two nonlinear systems. A comparison with two different adaptive modeling techniques, namely a dynamic evolving neural-fuzzy inference system (DENFIS) and neural networks trained with online backpropagation, highlights the advantages of the proposed methodology.
Segregation of tactile input features in neurons of the cuneate nucleus.
Jörntell, Henrik; Bengtsson, Fredrik; Geborek, Pontus; Spanne, Anton; Terekhov, Alexander V; Hayward, Vincent
2014-09-17
Our tactile perception of external objects depends on skin-object interactions. The mechanics of contact dictates the existence of fundamental spatiotemporal input features-contact initiation and cessation, slip, and rolling contact-that originate from the fact that solid objects do not interpenetrate. However, it is unknown whether these features are represented within the brain. We used a novel haptic interface to deliver such inputs to the glabrous skin of finger/digit pads and recorded from neurons of the cuneate nucleus (the brain's first level of tactile processing) in the cat. Surprisingly, despite having similar receptive fields and response properties, each cuneate neuron responded to a unique combination of these inputs. Hence, distinct haptic input features are encoded already at subcortical processing stages. This organization maps skin-object interactions into rich representations provided to higher cortical levels and may call for a re-evaluation of our current understanding of the brain's somatosensory systems.
Facial Feature Tracking and Head Pose Tracking as Input for Platform Games
Andersson, Anders Tobias
2016-01-01
Modern facial feature tracking techniques can automatically extract and accurately track multiple facial landmark points from faces in video streams in real time. Facial landmark points are deﬁned as points distributed on a face in regards to certain facial features, such as eye corners and face contour. This opens up for using facial feature movements as a handsfree human-computer interaction technique. These alternatives to traditional input devices can give a more interesting gaming experi...
Tao Hu
2013-01-01
Full Text Available Horizontal attenuation total reflection Fourier transformation infrared spectroscopy (HATR-FT-IR studies on cuscutae semen and its confusable varieties Japanese dodder and sinapis semen combined with discrete wavelet transformation (DWT and radial basis function (RBF neural networks have been conducted in order to classify them. DWT is used to decompose the FT-IRs of cuscutae semen, Japanese dodder, and sinapis semen. Two main scales are selected as the feature extracting space in the DWT domain. According to the distribution of cuscutae semen, Japanese dodder, and sinapis semen’s FT-IRs, three feature regions are determined at detail 3, and two feature regions are determined at detail 4 by selecting two scales in the DWT domain. Thus five feature parameters form the feature vector. The feature vector is input to the RBF neural networks to train so as to accurately classify the cuscutae semen, Japanese dodder, and sinapis semen. 120 sets of FT-IR data are used to train and test the proposed method, where 60 sets of data are used to train samples, and another 60 sets of FT-IR data are used to test samples. Experimental results show that the accurate recognition rate of cuscutae semen, Japanese dodder, and sinapis semen is average of 100.00%, 98.33%, and 100.00%, respectively, following the proposed method.
Deterministic System Identification Using RBF Networks
Joilson Batista de Almeida Rego
2014-01-01
Full Text Available This paper presents an artificial intelligence application using a nonconventional mathematical tool: the radial basis function (RBF networks, aiming to identify the current plant of an induction motor or other nonlinear systems. Here, the objective is to present the RBF response to different nonlinear systems and analyze the obtained results. A RBF network is trained and simulated in order to obtain the dynamical solution with basin of attraction and equilibrium point for known and unknown system and establish a relationship between these dynamical systems and the RBF response. On the basis of several examples, the results indicating the effectiveness of this approach are demonstrated.
Oliveira, Roberta B; Pereira, Aledir S; Tavares, João Manuel R S
2017-10-01
The number of deaths worldwide due to melanoma has risen in recent times, in part because melanoma is the most aggressive type of skin cancer. Computational systems have been developed to assist dermatologists in early diagnosis of skin cancer, or even to monitor skin lesions. However, there still remains a challenge to improve classifiers for the diagnosis of such skin lesions. The main objective of this article is to evaluate different ensemble classification models based on input feature manipulation to diagnose skin lesions. Input feature manipulation processes are based on feature subset selections from shape properties, colour variation and texture analysis to generate diversity for the ensemble models. Three subset selection models are presented here: (1) a subset selection model based on specific feature groups, (2) a correlation-based subset selection model, and (3) a subset selection model based on feature selection algorithms. Each ensemble classification model is generated using an optimum-path forest classifier and integrated with a majority voting strategy. The proposed models were applied on a set of 1104 dermoscopic images using a cross-validation procedure. The best results were obtained by the first ensemble classification model that generates a feature subset ensemble based on specific feature groups. The skin lesion diagnosis computational system achieved 94.3% accuracy, 91.8% sensitivity and 96.7% specificity. The input feature manipulation process based on specific feature subsets generated the greatest diversity for the ensemble classification model with very promising results. Copyright © 2017 Elsevier B.V. All rights reserved.
Kompella, Varun Raj; Schmidhuber, Juergen
2011-01-01
Slow Feature Analysis (SFA) extracts features representing the underlying causes of changes within a temporally coherent high-dimensional raw sensory input signal. Our novel incremental version of SFA (IncSFA) combines incremental Principal Components Analysis and Minor Components Analysis. Unlike standard batch-based SFA, IncSFA adapts along with non-stationary environments, is amenable to episodic training, is not corrupted by outliers, and is covariance-free. These properties make IncSFA a generally useful unsupervised preprocessor for autonomous learning agents and robots. In IncSFA, the CCIPCA and MCA updates take the form of Hebbian and anti-Hebbian updating, extending the biological plausibility of SFA. In both single node and deep network versions, IncSFA learns to encode its input streams (such as high-dimensional video) by informative slow features representing meaningful abstract environmental properties. It can handle cases where batch SFA fails.
Robustness of Input features from Noisy Silhouettes in Human Pose Estimation
Gong, Wenjuan; Fihl, Preben; Gonzàlez, Jordi;
2014-01-01
. In this paper, we explore this problem. First, We compare performances of several image features widely used for human pose estimation and explore their performances against each other and select one with best performance. Second, iterative closest point algorithm is introduced for a new quantitative...... measurement of noisy inputs. The proposed measurement is able to automatically discard noise, like camouflage from the background or shadows. With the proposed measurement, we split inputs into different noise levels and assess their pose estimation accuracies. Furthermore, we explore performances...
2001-01-01
The paper examines the role of input from a psychologicalperspective.By exploring the relation between language andthought,and the functions of memory,the paper aims to revealthat language,as a medium of thought,cannot be isolatedfrom thought in the thinking process.Therefore,input in thetarget language is to enable the learner to think in that language.Another idea borrowed from Psychology is the phenomenon offorgetting,which is resulted from interference.We argue thatproviding sufficient input for the learner is one of the effectiveways to minimize the degree of interference.The role of input isthen seen as the following:(1)fighting off mother tongueinterference;(2)internalizing L2 grammar;(3)defossilizingand maintaining interlanguage competence;(4)learningvocabulary in context.
Application of RBF Neural Network in OptimizingMachining Parameters
朱喜林; 吴博达; 武星星
2004-01-01
In machining processes, errors of rough in dimension, shape and location lead to changes in processing quantity, and the material of a workpiece may not be uniform. For these reasons, cutting force changes in machining, making the machining system deformable. Consequently errors in workpieces may occur. This is called the error reflection phenomenon. Generally, such errors can be reduced through repeated processing while using appropriate processing quantity in each processing based on operator's experience.According to the theory of error reflection, the error reflection coefficient indicates the extent to which errors of rough influence errors of workpieces. It is related to several factors such as machining condition, hardness of the workpiece, etc. This non-linear relation cannot be worked out using any formula. RBF neural network can approximate a non-linear function within any precision and be trained fast. In this paper, non-linear mapping ability of a fuzzy-neural network is utilized to approximate the non-linear relation. After training of the network with swatch collection obtained in experiments, an appropriate output can be obtained when an input is given. In this way, one can get the required number of processing and the processing quantity each time from the machining condition. Angular rigidity of a machining system,hardness of workpiece, etc., can be input in a form of fuzzy values. Feasibility in solving error reflection and optimizing machining parameters with a RBF neural network is verified by a simulation test with MATLAB.
Reinforcement learning on slow features of high-dimensional input streams.
Legenstein, Robert; Wilbert, Niko; Wiskott, Laurenz
2010-08-19
Humans and animals are able to learn complex behaviors based on a massive stream of sensory information from different modalities. Early animal studies have identified learning mechanisms that are based on reward and punishment such that animals tend to avoid actions that lead to punishment whereas rewarded actions are reinforced. However, most algorithms for reward-based learning are only applicable if the dimensionality of the state-space is sufficiently small or its structure is sufficiently simple. Therefore, the question arises how the problem of learning on high-dimensional data is solved in the brain. In this article, we propose a biologically plausible generic two-stage learning system that can directly be applied to raw high-dimensional input streams. The system is composed of a hierarchical slow feature analysis (SFA) network for preprocessing and a simple neural network on top that is trained based on rewards. We demonstrate by computer simulations that this generic architecture is able to learn quite demanding reinforcement learning tasks on high-dimensional visual input streams in a time that is comparable to the time needed when an explicit highly informative low-dimensional state-space representation is given instead of the high-dimensional visual input. The learning speed of the proposed architecture in a task similar to the Morris water maze task is comparable to that found in experimental studies with rats. This study thus supports the hypothesis that slowness learning is one important unsupervised learning principle utilized in the brain to form efficient state representations for behavioral learning.
Reinforcement learning on slow features of high-dimensional input streams.
Robert Legenstein
Full Text Available Humans and animals are able to learn complex behaviors based on a massive stream of sensory information from different modalities. Early animal studies have identified learning mechanisms that are based on reward and punishment such that animals tend to avoid actions that lead to punishment whereas rewarded actions are reinforced. However, most algorithms for reward-based learning are only applicable if the dimensionality of the state-space is sufficiently small or its structure is sufficiently simple. Therefore, the question arises how the problem of learning on high-dimensional data is solved in the brain. In this article, we propose a biologically plausible generic two-stage learning system that can directly be applied to raw high-dimensional input streams. The system is composed of a hierarchical slow feature analysis (SFA network for preprocessing and a simple neural network on top that is trained based on rewards. We demonstrate by computer simulations that this generic architecture is able to learn quite demanding reinforcement learning tasks on high-dimensional visual input streams in a time that is comparable to the time needed when an explicit highly informative low-dimensional state-space representation is given instead of the high-dimensional visual input. The learning speed of the proposed architecture in a task similar to the Morris water maze task is comparable to that found in experimental studies with rats. This study thus supports the hypothesis that slowness learning is one important unsupervised learning principle utilized in the brain to form efficient state representations for behavioral learning.
Multi-Input Converter with MPPT Feature for Wind-PV Power Generation System
Chih-Lung Shen
2013-01-01
Full Text Available A multi-input converter (MIC to process wind-PV power is proposed, designed, analyzed, simulated, and implemented. The MIC cannot only process solar energy but deal with wind power, of which structure is derived from forward-type DC/DC converter to step-down/up voltage for charger systems, DC distribution applications, or grid connection. The MIC comprises an upper modified double-ended forward, a lower modified double-ended forward, a common output inductor, and a DSP-based system controller. The two modified double-ended forwards can operate individually or simultaneously so as to accommodate the variation of the hybrid renewable energy under different atmospheric conditions. While the MIC operates at interleaving mode, better performance can be achieved and volume also is reduced. The proposed MIC is capable of recycling the energy stored in the leakage inductance and obtaining high step-up output voltage. In order to draw maximum power from wind turbine and PV panel, perturb-and-observe method is adopted to achieve maximum power point tracking (MPPT feature. The MIC is constructed, analyzed, simulated, and tested. Simulations and hardware measurements have demonstrated the feasibility and functionality of the proposed multi-input converter.
贾伟宽; 赵德安; 刘晓洋; 唐书萍; 阮承治; 姬伟
2015-01-01
In order to improve the recognition precision and speed for apple, and further improve the harvesting efficiency of apple harvesting robot, an apple recognition method based on combiningK-means clustering segmentation with genetic radial basis function (RBF) neural network is proposed. Firstly, the captured apple image is transformed into L*a*b* color space, and then under this color space, theK-means clustering algorithm is used to segment the apple image. The color feature components and shape components of segmented image are extracted respectively. The color features include R, G, B, H, S and I, a total of 6 feature components; and the shape features include circular variance, density, ratio of perimeter square to area, and 7 Hu invariant moments, a total of 10 shape components. These extracted 16 features are used as the inputs of neural network to train RBF neural network, and get the apple recognition model. Due to some inherent defects the RBF neural network has, such as low learning rate, easily causing over fitting phenomenon, genetic algorithm (GA) is introduced to optimize the connection weights and the number of hidden layer neurons. In this study, a new optimization way is adopted, that is, the hybrid encoding of the number of hidden layer neurons and connection weights is carried out simultaneously. This moment, the learning of weights is not completed, and the least mean square (LMS) is used to further learn the connection weights. Finally, an optimized neural network model (GA-RBF-LMS) is established, which is to improve the operating efficiency and recognition precision. In the experiments, there are 150 images captured, and they have 229 apples; among them 50 images are selected as training samples, and the rest as testing samples. Every image for training sample has only one apple, so the testing samples have 179 apples. In order to get the precise model, fruits of apple are together with branches and leaves for training during the training
RBF networks with mixed radial basis functions
Ciftcioglu, O.; Sariyildiz, I.S.
2000-01-01
After the introduction to neural network technology as multivariable function approximation, radial basis function (RBF) networks have been studied in many different aspects in recent years. From the theoretical viewpoint, approximation and uniqueness of the interpolation is studied and it has been
Performance Investigation of the RBF Localization Algorithm
Juraj Machaj
2013-01-01
Full Text Available In the present paper the impact of network properties on localization accuracy of Rank Based Fingerprinting algorithm will be investigated. Rank Based Fingerprinting (RBF will be described in detail together with Nearest Neighbour fingerprinting algorithms. RBF algorithm is a new algorithm and was designed as improvement of standard fingerprinting algorithms. Therefore exhaustive testing needs to be performed. This testing is mainly focused on optimal distribution of APs and its impact on positioning accuracy. Simulations were performed in Matlab environment in three different scenarios. In the first scenario different numbers of APs were implemented in the area to estimate the impact of APs number on the localization accuracy of the Rank Based Fingerprinting algorithm. The second scenario was introduced to evaluate the impact of APs placement in the localization area on the accuracy of the positioning using fingerprinting algorithms. The last scenario was proposed to investigate an impact of the number of heard APs and distribution of the RSS values on the accuracy of the RBF algorithm. Results achieved by the RBF algorithm in the first and second scenarios were compared to commonly used NN and WKNN algorithms.
A Regularized SNPOM for Stable Parameter Estimation of RBF-AR(X) Model.
Zeng, Xiaoyong; Peng, Hui; Zhou, Feng
2017-01-20
Recently, the radial basis function (RBF) network-style coefficients AutoRegressive (with exogenous inputs) [RBF-AR(X)] model identified by the structured nonlinear parameter optimization method (SNPOM) has attracted considerable interest because of its significant performance in nonlinear system modeling. However, this promising technique may occasionally confront the problem that the parameters are divergent in the optimization process, which may be a potential issue ignored by most researchers. In this paper, a regularized SNPOM, together with the regularization parameter detection technique, is presented to estimate the parameters of RBF-AR(X) models. This approach first separates the parameters of an RBF-AR(X) model into a linear parameters set and a nonlinear parameters set, and then combines a gradient-based nonlinear optimization algorithm for estimating the nonlinear parameters and the regularized least squares method for estimating the linear parameters. Several examples demonstrate that the proposed approach is effective to cope with the potential unstable problem in the parameters search process, and may also yield better or similar multistep forecasting accuracy and better robustness than the previous method.
The Ineffectiveness of the Provision of Input on the Problematic Grammatical Feature of Articles
Morgan, Gareth
2017-01-01
This study examined the value of giving specific input on the use of articles on an undergraduate English for Academic Purposes (EAP) course. This topic was chosen as previous cohorts had generated a noticeable amount of errors in their use of articles, and developing written grammatical accuracy was one of the course's aims. Participants were…
RBF neural network prediction on weak electrical signals in Aloe vera var. chinensis
Wang, Lanzhou; Zhao, Jiayin; Wang, Miao
2008-10-01
A Gaussian radial base function (RBF) neural network forecast on signals in the Aloe vera var. chinensis by the wavelet soft-threshold denoised as the time series and using the delayed input window chosen at 50, is set up to forecast backward. There was the maximum amplitude at 310.45μV, minimum -75.15μV, average value -2.69μV and Aloe vera var. chinensis respectively. The electrical signal in Aloe vera var. chinensis is a sort of weak, unstable and low frequency signals. A result showed that it is feasible to forecast plant electrical signals for the timing by the RBF. The forecast data can be used as the preferences for the intelligent autocontrol system based on the adaptive characteristic of plants to achieve the energy saving on the agricultural production in the plastic lookum or greenhouse.
Study of CNG/diesel dual fuel engine's emissions by means of RBF neural network
刘震涛; 费少梅
2004-01-01
Great efforts have been made to resolve the serious environmental pollution and inevitable declining of energy resources. A review of Chinese fuel reserves and engine technology showed that compressed natural gas (CNG)/diesel dual fuel engine (DFE) was one of the best solutions for the above problems at present. In order to study and improve the emission performance of CNG/diesel DFE, an emission model for DFE based on radial basis function (RBF) neural network was developed which was a black-box input-output training data model not require priori knowledge. The RBF centers and the connected weights could be selected automatically according to the distribution of the training data in input-output space and the given approximating error. Studies showed that the predicted results accorded well with the experimental data over a large range of operating conditions from low load to high load. The developed emissions model based on the RBF neural network could be used to successfully predict and optimize the emissions performance of DFE. And the effect of the DFE main performance parameters, such as rotation speed, load, pilot quantity and injection timing, were also predicted by means of this model. In resume, an emission prediction model for CNG/diesel DFE based on RBF neural network was built for analyzing the effect of the main performance parameters on the CO, NOx emissions of DFE. The predicted results agreed quite well with the traditional emissions model, which indicated that the model had certain application value, although it still has some limitations, because of its high dependence on the quantity of the experimental sample data.
Study of CNG/diesel dual fuel engine's emissions by means of RBF neural network.
Liu, Zhen-tao; Fei, Shao-mei
2004-08-01
Great efforts have been made to resolve the serious environmental pollution and inevitable declining of energy resources. A review of Chinese fuel reserves and engine technology showed that compressed natural gas (CNG)/diesel dual fuel engine (DFE) was one of the best solutions for the above problems at present. In order to study and improve the emission performance of CNG/diesel DFE, an emission model for DFE based on radial basis function (RBF) neural network was developed which was a black-box input-output training data model not require priori knowledge. The RBF centers and the connected weights could be selected automatically according to the distribution of the training data in input-output space and the given approximating error. Studies showed that the predicted results accorded well with the experimental data over a large range of operating conditions from low load to high load. The developed emissions model based on the RBF neural network could be used to successfully predict and optimize the emissions performance of DFE. And the effect of the DFEmain performance parameters, such as rotation speed, load, pilot quantity and injection timing, were also predicted by means of this model. In resumé, an emission prediction model for CNG/diesel DFE based on RBF neural network was built for analyzing the effect of the main performance parameters on the CO, NOx, emissions of DFE. The predicted results agreed quite well with the traditional emissions model, which indicated that the model had certain application value, although it still has some limitations, because of its high dependence on the quantity of the experimental sample data.
Meshless RBF based pseudospectral solution of acoustic wave equation
Mishra, Pankaj K
2015-01-01
Chebyshev pseudospectral (PS) methods are reported to provide highly accurate solution using polynomial approximation. Use of polynomial basis functions in PS algorithms limits the formulation to univariate systems constraining it to tensor product grids for multi-dimensions. Recent studies have shown that replacing the polynomial by radial basis functions in pseudospectral method (RBF-PS) has the advantage of using irregular grids for multivariate systems. A RBF-PS algorithm has been presented here for the numerical solution of inhomogeneous Helmholtz's equation using Gaussian RBF for derivative approximation. Efficacy of RBF approximated derivatives has been checked through error analysis comparison with PS method. Comparative study of PS, RBF-PS and finite difference approach for the solution of a linear boundary value problem has been performed. Finally, a typical frequency domain acoustic wave propagation problem has been solved using Dirichlet boundary condition and a point source. The algorithm present...
A survey on RBF Neural Network for Intrusion Detection System
Henali Sheth
2014-12-01
Full Text Available Network security is a hot burning issue nowadays. With the help of technology advancement intruders or hackers are adopting new methods to create different attacks in order to harm network security. Intrusion detection system (IDS is a kind of security software which inspects all incoming and outgoing network traffic and it will generate alerts if any attack or unusual behavior is found in a network. Various approaches are used for IDS such as data mining, neural network, genetic and statistical approach. Among this Neural Network is more suitable approach for IDS. This paper describes RBF neural network approach for Intrusion detection system. RBF is a feed forward and supervise technique of neural network.RBF approach has good classification ability but its performance depends on its parameters. Based on survey we find that RBF approach has some short comings. In order to overcome this we need to do proper optimization of RBF parameters.
王莉; 王德明; 张广明; 周献中
2011-01-01
A radical basis function (RBF) neural network model combined with rough sets was used to predict short-term wind speed. Rough sets were used to reduce input feature space so that the significant factors for wind speed prediction could be found as the input variables of RBF neural network prediction model. Online rolling optimization was adopted in training RBF neural network. The latest sample was added into the training sets, thus the prediction model could catch recent changes of wind speed. The proposed method was used to predict wind speed in 1 h. Simulation results showed that the method had advantages of simplicity and high precision.%结合粗糙集提出了一种RBF神经网络短期风速预测模型.采用粗糙集对预测模型的输入特征空间进行约简,找出对未来预测的风速具有主要影响的因素,以此作为RBF神经网络预测模型的输入变量；在RBF神经网络训练的过程中,采用在线滚动优化策略,将最新的样本加入训练集,从而使预测模型能够跟踪风速的最新变化.将提出的方法用于某风电场的lh短期风速预测,仿真实验结果表明该方法具有结构简单、预测精度高的优点.
Ting Wang
2015-01-01
Full Text Available Machine learning is a very important approach to pattern classification. This paper provides a better insight into Incremental Attribute Learning (IAL with further analysis as to why it can exhibit better performance than conventional batch training. IAL is a novel supervised machine learning strategy, which gradually trains features in one or more chunks. Previous research showed that IAL can obtain lower classification error rates than a conventional batch training approach. Yet the reason for that is still not very clear. In this study, the feasibility of IAL is verified by mathematical approaches. Moreover, experimental results derived by IAL neural networks on benchmarks also confirm the mathematical validation.
Compensation for unmatched uncertainty with adaptive RBF network
user
radial basis function (RBF) neural networks have showed strong universal approximation ability for unknown ..... w is the ideal constant weight, the ... w with the weight estimation error )(~ twi ..... Gaussian networks for direct adaptive control.
The overlapped radial basis function-finite difference (RBF-FD) method: A generalization of RBF-FD
Shankar, Varun
2017-08-01
We present a generalization of the RBF-FD method that computes RBF-FD weights in finite-sized neighborhoods around the centers of RBF-FD stencils by introducing an overlap parameter δ ∈ (0 , 1 ] such that δ = 1 recovers the standard RBF-FD method and δ = 0 results in a full decoupling of stencils. We provide experimental evidence to support this generalization, and develop an automatic stabilization procedure based on local Lebesgue functions for the stable selection of stencil weights over a wide range of δ values. We provide an a priori estimate for the speedup of our method over RBF-FD that serves as a good predictor for the true speedup. We apply our method to parabolic partial differential equations with time-dependent inhomogeneous boundary conditions - Neumann in 2D, and Dirichlet in 3D. Our results show that our method can achieve as high as a 60× speedup in 3D over existing RBF-FD methods in the task of forming differentiation matrices.
Ou, Yu-Yen; Gromiha, M Michael; Chen, Shu-An; Suwa, Makiko
2008-06-01
Discriminating outer membrane proteins (OMPs) from other folding types of globular and membrane proteins is an important task both for identifying OMPs from genomic sequences and for the successful prediction of their secondary and tertiary structures. We have developed a method based on radial basis function networks and position specific scoring matrix (PSSM) profiles generated by PSI-BLAST and non-redundant protein database. Our approach with PSSM profiles has correctly predicted the OMPs with a cross-validated accuracy of 96.4% in a set of 1251 proteins, which contain 206 OMPs, 667 globular proteins and 378 alpha-helical inner membrane proteins. Furthermore, we applied our method on a dataset containing 114 OMPs, 187 TMH proteins and 195 globular proteins obtained with less than 20% sequence identity and obtained the cross-validated accuracy of 95%. This accuracy of discriminating OMPs is higher than other methods in the literature and our method could be used as an effective tool for dissecting OMPs from genomic sequences. We have developed a prediction server, TMBETADISC-RBF, which is available at http://rbf.bioinfo.tw/~sachen/OMP.html.
Differentiation of digital tb images using texture analysis and rbf classifier.
Priya, E; Srinivasan, S; Ramakrishnan, S
2012-01-01
In this work, differentiation of positive and negative images of Tuberculosis (TB) sputum smear has been attempted using statistical method based on Gray Level Co-occurrence Matrix (GLCM). The sputum smear images (N=100) recorded under standard image acquisition protocol are considered for this work. Second order statistical texture analysis is performed on the acquired images using GLCM method and a set of nineteen features are derived. Principal Component Analysis (PCA) is then employed to reduce feature sets, to enhance the efficiency of differentiation and to reduce the redundancy. These feature sets are further classified using Radial Basis Function (RBF) classifier. Results show that GLCM is able to differentiate positive and negative TB images. Correlation is found to be high for many of the parameters. Application of PCA reduced the number of features to four which had maximum magnitude in the first principal component. Higher classification accuracy is achieved using RBF classifier. It appears that this method of texture analysis could be useful to develop automated system for characterization and classification of digital TB sputum smear images.
Identification of TSS in the Human Genome Based on a RBF Neural Network
Zhi-Hong Peng; Jie Chen; Li-Jun Cao; Ting-Ting Gao
2006-01-01
The identification of functional motifs in a DNA sequence is fundamentally a statistical pattern recognition problem. This paper introduces a new algorithm for the recognition of functional transcription start sites (TSSs) in human genome sequences, in which a RBF neural network is adopted, and an improved heuristic method for a 5-tuple feature viable construction, is proposed and implemented in two RBFPromoter and ImpRBFPromoter packages developed in Visual C++6.0. The algorithm is evaluated on several different test sequence sets. Compared with several other promoter recognition programs, this algorithm is proved to be more flexible, with stronger learning ability and higher accuracy.
Study of CNG／diesel dual fuel engine＇s emissions by means of RBF neural network
刘震涛; 费少梅
2004-01-01
Great efforts have been made to resolve the serious environmental pollution and inevitable declining of energy resources. A review of Chinese fuel reserves and engine technology showed that compressed natural gas (CNG)/diesel dual fuel engine (DFE) was one of the best solutions for the above problems at present. In order to study and improve the emission performance of CNG/diesel DFE, an emission model for DFE based on radial basis function (RBF) neural network was developed which was a black-box input-output training data model not require priori knowledge. The RBF centers and the connected weights could be selected automatically according to the distribution of the training data in input-output space and the given approximating error. Studies showed that the predicted results accorded well with the experimental data over a large range of operating conditions from low load to high load. The developed emissions model based on the RBF neural network could be used to successfully predict and optimize the emissions performance of DFE. And the effect of the DFE main performance parameters, such as rotation speed, load, pilot quantity and injection timing, were also predicted by means of this model. In resum6, an emission prediction model for CNG/diesel DFE based on RBF neural network was built for analyzing the effect of the main performance parameters on the CO, NOx emissions of DFE. The predicted results agreed quite well with the traditional emissions model, which indicated that the model had certain application value, although it still has some limitations, because of its high dependence on the quantity of the experimental sample data.
Rishi, Rahul; Choudhary, Amit; Singh, Ravinder; Dhaka, Vijaypal Singh; Ahlawat, Savita; Rao, Mukta
2010-02-01
In this paper we propose a system for classification problem of handwritten text. The system is composed of preprocessing module, supervised learning module and recognition module on a very broad level. The preprocessing module digitizes the documents and extracts features (tangent values) for each character. The radial basis function network is used in the learning and recognition modules. The objective is to analyze and improve the performance of Multi Layer Perceptron (MLP) using RBF transfer functions over Logarithmic Sigmoid Function. The results of 35 experiments indicate that the Feed Forward MLP performs accurately and exhaustively with RBF. With the change in weight update mechanism and feature-drawn preprocessing module, the proposed system is competent with good recognition show.
MIMO Lyapunov Theory-Based RBF Neural Classifier for Traffic Sign Recognition
King Hann Lim
2012-01-01
Full Text Available Lyapunov theory-based radial basis function neural network (RBFNN is developed for traffic sign recognition in this paper to perform multiple inputs multiple outputs (MIMO classification. Multidimensional input is inserted into RBF nodes and these nodes are linked with multiple weights. An iterative weight adaptation scheme is hence designed with regards to the Lyapunov stability theory to obtain a set of optimum weights. In the design, the Lyapunov function has to be well selected to construct an energy space with a single global minimum. Weight gain is formed later to obey the Lyapunov stability theory. Detail analysis and discussion on the proposed classifier’s properties are included in the paper. The performance comparisons between the proposed classifier and some existing conventional techniques are evaluated using traffic sign patterns. Simulation results reveal that our proposed system achieved better performance with lower number of training iterations.
Radiation acquisition and RBF neural network analysis on BOF end-point control
Zhao, Qi; Wen, Hong-yuan; Zhou, Mu-chun; Chen, Yan-ru
2008-12-01
There are some problems in Basic Oxygen Furnace (BOF) steelmaking end-point control technology at present. A new BOF end-point control model was designed, which was based on the character of carbon oxygen reaction in Basic Oxygen Furnace steelmaking process. The image capture and transformation system was established by Video for Windows (VFW) library function, which is a video software development package promoted by Microsoft Corporation. In this paper, the Radial Basic Function (RBF) neural network model was established by using the real-time acquisition information. The input parameters can acquire easily online and the output parameter is the end-point time, which can compare with the actual value conveniently. The experience results show that the predication result is ideal and the experiment results show the model can work well in the steelmaking adverse environment.
An adaptive control for a variable speed wind turbine using RBF neural network
El Mjabber E.
2016-01-01
Full Text Available In this work, a controller based on Radial Basis Functions (RBF for network adaptation is considered. The adaptive Neural Network (NN control approximates the nonlinear dynamics of the wind turbine based on input/output measurement and ensures smooth tracking of optimal tip speed-ratio at different wind speeds. The wind turbine system and this controller were modeled and a program to integrate the obtained coupled equations was developed under Matlab/Simulink software package. Then, performance of the controller was studied numerically. The proposed controller was found to effectively improve the control performance against large uncertainty of the wind turbine system. comparison with nonlinear dynamic State feedback control with Kalman filter controller was performed, and the obtained results have demonstrated the relevance of this RBFNN based controller.
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.
Application of BP NN and RBF NN in Modeling Activated Sludge System
王维斌; 郑丕谔; 李金勇
2003-01-01
Based on the operation data from a certain wastewater treatment plant(WWTP) in northeast China, the models of back propagation neural network(BP NN) and radial basis function neural network(RBF NN) have been designed respectively and the ability of convergence and generalization has been analyzed separately. As for BP NN, the effects of numbers of layers and nodes have been studied; as for RBF NN, the influences of the number of nodes and the RBF′s width have been studied. It is concluded that BP NN has converged much slowly in comparison with RBF NN. The conclusion that the RBF NN is suitable for modeling activated sludge system has been drawn. An automatically optimum design program for RBF NN has been developed, through which the RBF NN model of traditional activated sludge system has been established.
A RBF Network Learning Scheme Using Immune Algorithm Based on Information Entropy
GONG Xin-bao; ZANG Xiao-gang; ZHOU Xi-lang
2005-01-01
A hybrid learning method combining immune algorithm and least square method is proposed to design the radial basis function(RBF) networks. The immune algorithm based on information entropy is used to determine the structure and parameters of RBF nonlinear hidden layer, and weights of RBF linear output layer are computed with least square method. By introducing the diversity control and immune memory mechanism, the algorithm improves the efficiency and overcomes the immature problem in genetic algorithm. Computer simulations demonstrate that the RBF networks designed in this method have fast convergence speed with good performances.
NO mediates downregulation of RBF after a prolonged reduction of renal perfusion pressure in SHR
Sørensen, Charlotte Mehlin; Leyssac, Paul Peter; Skott, Ole
2003-01-01
The aim of the study was to investigate mechanisms underlying the downregulation of renal blood flow (RBF) after a prolonged reduction in renal perfusion pressure (RPP) in adult spontaneously hypertensive rats (SHR). We tested the effect on the RBF response of clamping plasma ANG II in sevoflurane...... of plasma ANG II concentrations, general COX inhibition, and specific inhibition of COX-2. In contrast, clamping the NO system diminished the ability of SHR to downregulate RBF to a lower level. The downregulation of RBF was not associated with a resetting of the lower limit of autoregulation in the control...... of vasoconstrictory prostaglandins....
Study of RBF Nerve Network Tuning PD Control Algoritm of Bilateral Servo System
Guang Wen
2013-01-01
Full Text Available In construction tele-robot system. When p-f architecture force feedback was used, the impact of large feedback force result in the strike-like feeling on the operators hand. If the amplitude is high, it will cause the control unstable. So a improved force feedback control method with the feature of a T-S fuzzy feedback coefficient, which could be modified online nonlinearly and continuously, is developed. A RBF-PID force controller is also designed, and formed a bilateral hydraulic servo control system. The experimental results indicate that the new improved control method reduced the impact of the feedback force, enhanced the compliance and transparency of the tele-operation of construction tele-robot system.
RBF Nerve Network Tuning PD Control Scheme for Tele-operation Robot Servo System
Guang Wen
2013-11-01
Full Text Available In the bilateral hydraulic servo control system of a construction tele-robot with in-situ force sensing, the p-f type force feedback architecture is liable to result in an impact on the operator hand, and its high amplitude will cause the control unstable. In order to solve this problem an improved force feedback control method with the feature of a T-S fuzzy feedback coefficient, which could be modified online nonlinearly and continuously, is proposed. And a RBF-PID force controller is also designed, and formed a bilateral hydraulic servo control system. The experimental results indicate that the new improved control method reduced the impact of the feedback force, the compliance and transparency of the tele-operation of construction tele-robot system are enhanced.
Weber, E.; Rosenauer, M.; Verhaert, P.D.E.M.; Vellekoop, M.J.
2010-01-01
We present an optofluidic microsystem integrated onto a single device featuring on-chip light guiding and positioning under stable and low-loss conditions. Integration of optical components onto a microfluidic chip offers numerous new possibilities in the field of particle and cell analysis, but rec
Classification data mining method based on dynamic RBF neural networks
Zhou, Lijuan; Xu, Min; Zhang, Zhang; Duan, Luping
2009-04-01
With the widely application of databases and sharp development of Internet, The capacity of utilizing information technology to manufacture and collect data has improved greatly. It is an urgent problem to mine useful information or knowledge from large databases or data warehouses. Therefore, data mining technology is developed rapidly to meet the need. But DM (data mining) often faces so much data which is noisy, disorder and nonlinear. Fortunately, ANN (Artificial Neural Network) is suitable to solve the before-mentioned problems of DM because ANN has such merits as good robustness, adaptability, parallel-disposal, distributing-memory and high tolerating-error. This paper gives a detailed discussion about the application of ANN method used in DM based on the analysis of all kinds of data mining technology, and especially lays stress on the classification Data Mining based on RBF neural networks. Pattern classification is an important part of the RBF neural network application. Under on-line environment, the training dataset is variable, so the batch learning algorithm (e.g. OLS) which will generate plenty of unnecessary retraining has a lower efficiency. This paper deduces an incremental learning algorithm (ILA) from the gradient descend algorithm to improve the bottleneck. ILA can adaptively adjust parameters of RBF networks driven by minimizing the error cost, without any redundant retraining. Using the method proposed in this paper, an on-line classification system was constructed to resolve the IRIS classification problem. Experiment results show the algorithm has fast convergence rate and excellent on-line classification performance.
Implementation of pattern recognition algorithm based on RBF neural network
Bouchoux, Sophie; Brost, Vincent; Yang, Fan; Grapin, Jean Claude; Paindavoine, Michel
2002-12-01
In this paper, we present implementations of a pattern recognition algorithm which uses a RBF (Radial Basis Function) neural network. Our aim is to elaborate a quite efficient system which realizes real time faces tracking and identity verification in natural video sequences. Hardware implementations have been realized on an embedded system developed by our laboratory. This system is based on a DSP (Digital Signal Processor) TMS320C6x. The optimization of implementations allow us to obtain a processing speed of 4.8 images (240x320 pixels) per second with a correct rate of 95% of faces tracking and identity verification.
Normalized RBF networks: application to a system of integral equations
Golbabai, A; Seifollahi, S; Javidi, M [Department of Mathematics, Iran University of Science and Technology, Narmak, Tehran 16844 (Iran, Islamic Republic of)], E-mail: golbabai@iust.ac.ir, E-mail: seif@iust.ac.ir, E-mail: mojavidi@yahoo.com
2008-07-15
Linear integral and integro-differential equations of Fredholm and Volterra types have been successfully treated using radial basis function (RBF) networks in previous works. This paper deals with the case of a system of integral equations of Fredholm and Volterra types with a normalized radial basis function (NRBF) network. A novel learning algorithm is developed for the training of NRBF networks in which the BFGS backpropagation (BFGS-BP) least-squares optimization method as a recursive model is used. In the approach presented here, a trial solution is given by an NRBF network of incremental architecture with a set of unknown parameters. Detailed learning algorithms and concrete examples are also included.
Hájos, Norbert; Karlócai, Mária R; Németh, Beáta; Ulbert, István; Monyer, Hannah; Szabó, Gábor; Erdélyi, Ferenc; Freund, Tamás F; Gulyás, Attila I
2013-07-10
Hippocampal sharp waves and the associated ripple oscillations (SWRs) are implicated in memory processes. These network events emerge intrinsically in the CA3 network. To understand cellular interactions that generate SWRs, we detected first spiking activity followed by recording of synaptic currents in distinct types of anatomically identified CA3 neurons during SWRs that occurred spontaneously in mouse hippocampal slices. We observed that the vast majority of interneurons fired during SWRs, whereas only a small portion of pyramidal cells was found to spike. There were substantial differences in the firing behavior among interneuron groups; parvalbumin-expressing basket cells were one of the most active GABAergic cells during SWRs, whereas ivy cells were silent. Analysis of the synaptic currents during SWRs uncovered that the dominant synaptic input to the pyramidal cell was inhibitory, whereas spiking interneurons received larger synaptic excitation than inhibition. The discharge of all interneurons was primarily determined by the magnitude and the timing of synaptic excitation. Strikingly, we observed that the temporal structure of synaptic excitation and inhibition during SWRs significantly differed between parvalbumin-containing basket cells, axoaxonic cells, and type 1 cannabinoid receptor (CB1)-expressing basket cells, which might explain their distinct recruitment to these synchronous events. Our data support the hypothesis that the active current sources restricted to the stratum pyramidale during SWRs originate from the synaptic output of parvalbumin-expressing basket cells. Thus, in addition to gamma oscillation, these GABAergic cells play a central role in SWR generation.
Shah, Syed Muhammad Saqlain; Batool, Safeera; Khan, Imran; Ashraf, Muhammad Usman; Abbas, Syed Hussnain; Hussain, Syed Adnan
2017-09-01
Automatic diagnosis of human diseases are mostly achieved through decision support systems. The performance of these systems is mainly dependent on the selection of the most relevant features. This becomes harder when the dataset contains missing values for the different features. Probabilistic Principal Component Analysis (PPCA) has reputation to deal with the problem of missing values of attributes. This research presents a methodology which uses the results of medical tests as input, extracts a reduced dimensional feature subset and provides diagnosis of heart disease. The proposed methodology extracts high impact features in new projection by using Probabilistic Principal Component Analysis (PPCA). PPCA extracts projection vectors which contribute in highest covariance and these projection vectors are used to reduce feature dimension. The selection of projection vectors is done through Parallel Analysis (PA). The feature subset with the reduced dimension is provided to radial basis function (RBF) kernel based Support Vector Machines (SVM). The RBF based SVM serves the purpose of classification into two categories i.e., Heart Patient (HP) and Normal Subject (NS). The proposed methodology is evaluated through accuracy, specificity and sensitivity over the three datasets of UCI i.e., Cleveland, Switzerland and Hungarian. The statistical results achieved through the proposed technique are presented in comparison to the existing research showing its impact. The proposed technique achieved an accuracy of 82.18%, 85.82% and 91.30% for Cleveland, Hungarian and Switzerland dataset respectively.
Using the Correlation Criterion to Position and Shape RBF Units for Incremental Modelling
Xun-Xian Wang; Sheng Chen; Chris J. Harris
2006-01-01
A novel technique is proposed for the incremental construction of sparse radial basis function (RBF) networks.The correlation between an RBF regressor and the training data is used as the criterion to position and shape the RBF node, and it is shown that this is equivalent to incrementally minimise the modelling mean square error. A guided random search optimisation method, called the repeated weighted boosting search, is adopted to append RBF nodes one by one in an incremental regression modelling procedure. The experimental results obtained using the proposed method demonstrate that it provides a viable alternative to the existing state-of-the-art modelling techniques for constructing parsimonious RBF models that generalise well.
Fatma Zohra Chelali
2015-01-01
Full Text Available Face recognition has received a great attention from a lot of researchers in computer vision, pattern recognition, and human machine computer interfaces in recent years. Designing a face recognition system is a complex task due to the wide variety of illumination, pose, and facial expression. A lot of approaches have been developed to find the optimal space in which face feature descriptors are well distinguished and separated. Face representation using Gabor features and discrete wavelet has attracted considerable attention in computer vision and image processing. We describe in this paper a face recognition system using artificial neural networks like multilayer perceptron (MLP and radial basis function (RBF where Gabor and discrete wavelet based feature extraction methods are proposed for the extraction of features from facial images using two facial databases: the ORL and computer vision. Good recognition rate was obtained using Gabor and DWT parameterization with MLP classifier applied for computer vision dataset.
Logarithmic Spiral-based Construction of RBF Classifiers
Mohamed Wajih Guerfala
2017-02-01
Full Text Available Clustering process is defined as grouping similar objects together into homogeneous groups or clusters. Objects that belong to one cluster should be very similar to each other, but objects in different clusters will be dissimilar. It aims to simplify the representation of the initial data. The automatic classification recovers all the methods allowing the automatic construction of such groups. This paper describes the design of radial basis function (RBF neural classifiers using a new algorithm for characterizing the hidden layer structure. This algorithm, called k-means Mahalanobis distance, groups the training data class by class in order to calculate the optimal number of clusters of the hidden layer, using two validity indexes. To initialize the initial clusters of k-means algorithm, the method of logarithmic spiral golden angle has been used. Two real data sets (Iris and Wine are considered to improve the efficiency of the proposed approach and the obtained results are compared with basic literature classifier
Efficient and exact mesh deformation using multiscale RBF interpolation
Kedward, L.; Allen, C. B.; Rendall, T. C. S.
2017-09-01
Radial basis function (RBF) interpolation is popular for mesh deformation due to robustness and generality, but the cost scales with the number of surface points sourcing the deformation as O (Ns3). Hence, there have been numerous works investigating efficient methods using reduced datasets. However, although reduced-data methods are efficient, they require a secondary method to treat an error vector field to ensure surface points not included in the primary deformation are moved to the correct location, and the volume mesh moved accordingly. A new method is presented which captures global and local motions at multiple scales using all the surface points, and so no correction stage is required; all surface points are used and a single interpolation built, but the cost and conditioning issues associated with RBF methods are eliminated. Moreover, the sparsity introduced is exploited using a wall distance function, to further reduce the cost. The method is compared to an efficient greedy method, and it is shown mesh quality is always comparable with or better than with the greedy method, and cost is comparable or cheaper at all stages. Surface mesh preprocessing is the dominant cost for reduced-data methods and this cost is reduced significantly here: greedy methods select points to minimise interpolation error, requiring repeated system solution and cost O (Nred4) to select Nred points; the multiscale method has no error, and the problem is transferred to a geometric search, with cost O (Ns log (Ns)), resulting in an eight orders of magnitude cost reduction for three-dimensional meshes. Furthermore, since the method is dependent on geometry, not deformation, it only needs to be applied once, prior to simulation, as the mesh deformation is decoupled from the point selection process.
Abghari, H.; van de Giesen, N.; Mahdavi, M.; Salajegheh, A.
2009-04-01
Artificial intelligence modeling of nonstationary rainfall-runoff has some restrictions in simulation accuracy due to the complexity and nonlinearity of training patterns. Preprocessing of trainings dataset could determine homogeneity of rainfall-runoff patterns before modeling. In this presentation, a new hybrid model of Artificial Intelligence in conjunction with clustering is introduced and applied to flow prediction. Simulation of Nazloochaei river flow in North-West Iran was the case used for development of a PNN-RBF model. PNN classify a training dataset in two groups based on Parezen theory using unsupervised classification. Subsequently each data group is used to train and test two RBF networks and the results are compared to the application of all data in a RBF network without classification. Results show that classification of rainfall-runoff patterns using PNN and prediction of runoff with RBF increase prediction precise of networks. Keywords: Probabilistic Neural Network, Radial Base Function Neural Network, Parezen theory, River Flow Prediction
Optimal design of structures for earthquake loads by a hybrid RBF-BPSO method
Eysa Salajegheh; Saeed Gholizadeh; Mohsen Khatibina
2008-01-01
The optimal seismic design of structures requires that time history analyses (THA) be carried out repeatedly. This makes the optimal design process inefficient, in particular, if an evolutionary algorithm is used. To reduce the overall time required for structural optimization, two artificial intelligence strategies are employed. In the first strategy, radial basis function (RBF) neural networks are used to predict the time history responses of structures in the optimization flow. In the second strategy, a binary particle swarm optimization (BPSO) is used to find the optimum design. Combining the RBF and BPSO, a hybrid RBF-BPSO optimization method is proposed in this paper, which achieves fast optimization with high computational performance. Two examples are presented and compared to determine the optimal weight of structures under earthquake loadings using both exact and approximate analyses. The numerical results demonstrate the computational advantages and effectiveness of the proposed hybrid RBF-BPSO optimization method for the seismic design of structures.
IMMUNE RBF NETWORK AND ITS APPLICATION IN THE MODULATION-STYLE RECOGNITION OF RADAR SIGNALS
Gong Xinbao; Zang Xiaogang; Zhou Xilang; Hu Guangrui
2003-01-01
Based on Immune Programming(IP), a novel Radial Basis Function (RBF) networkdesigning method is proposed. Through extracting the preliminary knowledge about the widthof the basis function as the vaccine to form the immune operator, the algorithm reduces thesearching space of canonical algorithm and improves the convergence speed. The application ofthe RBF network trained with the algorithm in the modulation-style recognition of radar signalsdemonstrates that the network has a fast convergence speed with good performances.
A Sliding Mode Control-based on a RBF Neural Network for Deburring Industry Robotic Systems
Yong Tao; Jiaqi Zheng; Yuanchang Lin
2016-01-01
A sliding mode control method based on radial basis function (RBF) neural network is proposed for the deburring of industry robotic systems. First, a dynamic model for deburring the robot system is established. Then, a conventional SMC scheme is introduced for the joint position tracking of robot manipulators. The RBF neural network based sliding mode control (RBFNN-SMC) has the ability to learn uncertain control actions. In the RBFNN-SMC scheme, the adaptive tuning algorithms for network par...
铁路扣件图像检测中的RBF-SVM模型优化%Optimization of RBF-SVM model in railway fastener detection system
刘甲甲; 王凯; 袁建英; 江晓亮; 李柏林
2014-01-01
在开发的铁路扣件检测系统中，RBF-SVM被作为扣件图像分类识别的分类器。核参数的选择是RBF-SVM模型优化研究中的重要问题，将量子粒子群算法应用于参数的优化选择，在(c，γ)参数可调范围内产生初始种群，将种群中的个体作为RBF-SVM的参数进行学习；经过多次迭代获得最佳参数对(c，γ)，并将该参数对作为RBF-SVM的核参数训练支持向量机。实验表明，QPSO的性能优于传统的 PSO算法，该方法在解决支持向量机优化方面表现出了高效的收敛性和稳定性，并且在该方法的基础上形成的铁路扣件检测算法是切实可行的。%In the railway fastener detection system, RBF-SVM is used as image classifier for railway fasteners. The selection of kernel parameters is an important problem in RBF-SVM research. A parameter selection method based on quantum genet-ic algorithm(QPSO)is presented. Initial population is produced in the adjustable range of parameters c and γ, and individuals in it are used as the parameters of RBF-SVM to calculation; then by multi-iterations, the parameters (c,γ) are obtained which are corresponding to fitness of population, and used as kernel parameters of Radial Basis kernel Function of Support Vector Machine(RBF-SVM)to training model. The experimental results indicate that the QPSO algorithm outperforms PSO algorithm. It has a high convergence and stability, and the detection algorithm of rail fastener based on it is practicable.
An empirical RBF model of the magnetosphere parameterized by interplanetary and ground-based drivers
Tsyganenko, N. A.; Andreeva, V. A.
2016-11-01
In our recent paper (Andreeva and Tsyganenko, 2016), a novel method was proposed to model the magnetosphere directly from spacecraft data, with no a priori knowledge nor ad hoc assumptions about the geometry of the magnetic field sources. The idea was to split the field into the toroidal and poloidal parts and then expand each part into a weighted sum of radial basis functions (RBF). In the present work we take the next step forward by having developed a full-fledged model of the near magnetosphere, based on a multiyear set of space magnetometer data (1995-2015) and driven by ground-based and interplanetary input parameters. The model consolidates the largest ever amount of data and has been found to provide the best ever merit parameters, in terms of both the overall RMS residual field and record-high correlation coefficients between the observed and model field components. By experimenting with different combinations of input parameters and their time-averaging intervals, we found the best so far results to be given by the ram pressure Pd, SYM-H, and N-index by Newell et al. (2007). In addition, the IMF By has also been included as a model driver, with a goal to more accurately represent the IMF penetration effects. The model faithfully reproduces both externally and internally induced variations in the global distribution of the geomagnetic field and electric currents. Stronger solar wind driving results in a deepening of the equatorial field depression and a dramatic increase of its dawn-dusk asymmetry. The Earth's dipole tilt causes a consistent deformation of the magnetotail current sheet and a significant north-south asymmetry of the polar cusp depressions on the dayside. Next steps to further develop the new approach are also discussed.
A Hybrid Framework using RBF and SVM for Direct Marketing
M. Govidarajan
2013-05-01
Full Text Available one of the major developments in machine learning in the past decade is the ensemble method, which finds highly accurate classifier by combining many moderately accurate component classifiers. This paper addresses using an ensemble of classification methods for direct marketing. Direct marketing has become an important application field for data mining. In direct marketing, companies or organizations try to establish and maintain a direct relationship with their customers in order to target them individually for specific product offers or for fund raising. A variety of techniques have been employed for analysis ranging from traditional statistical methods to data mining approaches. In this research work, new hybrid classification method is proposed by combining classifiers in a heterogeneous environment using arcing classifier and their performances are analyzed in terms of accuracy. A Classifier ensemble is designed using Radial Basis Function (RBF and Support Vector Machine (SVM as base classifiers. Here, modified training sets are formed by resampling from original training set; classifiers constructed using these training sets and then combined by voting. Empirical results illustrate that the proposed hybrid systems provide more accurate direct marketing system.
Clavier, Amandine; Ruby, Vincent; Rincheval-Arnold, Aurore; Mignotte, Bernard; Guénal, Isabelle
2015-09-01
In accordance with its tumor suppressor role, the retinoblastoma protein pRb can ensure pro-apoptotic functions. Rbf1, the Drosophila homolog of Rb, also displays a pro-apoptotic activity in proliferative cells. We have previously shown that the Rbf1 pro-apoptotic activity depends on its ability to decrease the level of anti-apoptotic proteins such as the Bcl-2 family protein Buffy. Buffy often acts in an opposite manner to Debcl, the other Drosophila Bcl-2-family protein. Both proteins can localize at the mitochondrion, but the way they control apoptosis still remains unclear. Here, we demonstrate that Debcl and the pro-fission gene Drp1 are necessary downstream of Buffy to trigger a mitochondrial fragmentation during Rbf1-induced apoptosis. Interestingly, Rbf1-induced apoptosis leads to a Debcl- and Drp1-dependent reactive oxygen species production, which in turn activates the Jun Kinase pathway to trigger cell death. Moreover, we show that Debcl and Drp1 can interact and that Buffy inhibits this interaction. Notably, Debcl modulates Drp1 mitochondrial localization during apoptosis. These results provide a mechanism by which Drosophila Bcl-2 family proteins can control apoptosis, and shed light on a link between Rbf1 and mitochondrial dynamics in vivo. © 2015. Published by The Company of Biologists Ltd.
On the role of polynomials in RBF-FD approximations: I. Interpolation and accuracy
Flyer, Natasha; Fornberg, Bengt; Bayona, Victor; Barnett, Gregory A.
2016-09-01
Radial basis function-generated finite difference (RBF-FD) approximations generalize classical grid-based finite differences (FD) from lattice-based to scattered node layouts. This greatly increases the geometric flexibility of the discretizations and makes it easier to carry out local refinement in critical areas. Many different types of radial functions have been considered in this RBF-FD context. In this study, we find that (i) polyharmonic splines (PHS) in conjunction with supplementary polynomials provide a very simple way to defeat stagnation (also known as saturation) error and (ii) give particularly good accuracy for the tasks of interpolation and derivative approximations without the hassle of determining a shape parameter. In follow-up studies, we will focus on how to best use these hybrid RBF polynomial bases for FD approximations in the contexts of solving elliptic and hyperbolic type PDEs.
CLASSIFICATIONS OF EEG SIGNALS FOR MENTAL TASKS USING ADAPTIVE RBF NETWORK
薛建中; 郑崇勋; 闫相国
2004-01-01
Objective This paper presents classifications of mental tasks based on EEG signals using an adaptive Radial Basis Function (RBF) network with optimal centers and widths for the Brain-Computer Interface (BCI) schemes. Methods Initial centers and widths of the network are selected by a cluster estimation method based on the distribution of the training set. Using a conjugate gradient descent method, they are optimized during training phase according to a regularized error function considering the influence of their changes to output values. Results The optimizing process improves the performance of RBF network, and its best cognition rate of three task pairs over four subjects achieves 87.0%. Moreover, this network runs fast due to the fewer hidden layer neurons. Conclusion The adaptive RBF network with optimal centers and widths has high recognition rate and runs fast. It may be a promising classifier for on-line BCI scheme.
Erkan Beşdok
2009-08-01
Full Text Available This paper introduces a comparison of training algorithms of radial basis function (RBF neural networks for classification purposes. RBF networks provide effective solutions in many science and engineering fields. They are especially popular in the pattern classification and signal processing areas. Several algorithms have been proposed for training RBF networks. The Artificial Bee Colony (ABC algorithm is a new, very simple and robust population based optimization algorithm that is inspired by the intelligent behavior of honey bee swarms. The training performance of the ABC algorithm is compared with the Genetic algorithm, Kalman filtering algorithm and gradient descent algorithm. In the experiments, not only well known classification problems from the UCI repository such as the Iris, Wine and Glass datasets have been used, but also an experimental setup is designed and inertial sensor based terrain classification for autonomous ground vehicles was also achieved. Experimental results show that the use of the ABC algorithm results in better learning than those of others.
On the role of polynomials in RBF-FD approximations: II. Numerical solution of elliptic PDEs
Bayona, Victor; Flyer, Natasha; Fornberg, Bengt; Barnett, Gregory A.
2017-03-01
RBF-generated finite differences (RBF-FD) have in the last decade emerged as a very powerful and flexible numerical approach for solving a wide range of PDEs. We find in the present study that combining polyharmonic splines (PHS) with multivariate polynomials offers an outstanding combination of simplicity, accuracy, and geometric flexibility when solving elliptic equations in irregular (or regular) regions. In particular, the drawbacks on accuracy and stability due to Runge's phenomenon are overcome once the RBF stencils exceed a certain size due to an underlying minimization property. Test problems include the classical 2-D driven cavity, and also a 3-D global electric circuit problem with the earth's irregular topography as its bottom boundary. The results we find are fully consistent with previous results for data interpolation.
谭昶; 肖南峰
2011-01-01
针对手势识别的手区域分割、手势特征提取和手势分类的三个过程,提出了一种新的静态手势识别方法.改进了传统的RCE神经网络用于手区域的分割,具有更高的运行速度和更强的抗噪能力.依Freeman链码方向提取手的边缘到掌心的距离作为手势的特征向量.将上一步得到的手势特征向量作为RBF神经网络的输入,进行网络的训练和分类.实验验证了该方法的有效性和可行性,并用其实现了人和仿人机器人的剪刀石头布的猜拳游戏.%Hand gesture recognition usually includes hand image segmentation,features extraction,and hand gestures classification. In this paper,a new method is proposed to deal with the three phases of static hand gesture recognition. The traditional RCE neural network is improved and it is applied to hand image segmentation.The improved RCE neural network is proved to has running fast and strong ability of anti-noise. Freeman chain code is used to extract the distance from hand edge to the centre of the palm as feature vectors. Those feature vectors are used as the input of RBF neural network. Experiment resuits show this method is efficient and feasible. A scissors-paper-stone game between human and humanoid robot is developed by using this method.
An artificial Radial Basis Function (RBF) neural network model was developed for the prediction of mass transfer of the phospholipids from canola meal in supercritical CO2 fluid. The RBF kind of artificial neural networks (ANN) with orthogonal least squares (OLS) learning algorithm were used for mod...
RBF multiuser detector with channel estimation capability in a synchronous MC-CDMA system.
Ko, K; Choi, S; Kang, C; Hong, D
2001-01-01
The authors propose a multiuser detector with channel estimation capability using a radial basis function (RBF) network in a synchronous multicarrier-code division multiple access (MC-CDMA) system. The authors propose to connect an RBF network to the frequency domain to effectively utilize the frequency diversity. Simulations were performed over frequency-selective and multi-path fading channels. These simulations confirmed that the proposed receiver can be used both for the channel estimation and as a multi-user receiver, thus permitting an increase in the number of active users.
On-line identification of hybrid systems using an adaptive growing and pruning RBF neural network
Alizadeh, Tohid
2008-01-01
This paper introduces an adaptive growing and pruning radial basis function (GAP-RBF) neural network for on-line identification of hybrid systems. The main idea is to identify a global nonlinear model that can predict the continuous outputs of hybrid systems. In the proposed approach, GAP......-RBF neural network uses a modified unscented kalman filter (UKF) with forgetting factor scheme as the required on-line learning algorithm. The effectiveness of the resulting identification approach is tested and evaluated on a simulated benchmark hybrid system....
A Model to Predict Rolling Force of Finishing Stands with RBF Neural Networks
无
2005-01-01
In view of intrinsic imperfection of traditional models of rolling force, in order to improve the prediction accuracy of rolling force, a new method combining radial basis function (RBF) neural networks with traditional models to predict rolling force was proposed. The off-line simulation indicates that the predicted results are much more accurate than that with traditional models.
RBF-Type Artificial Neural Network Model Applied in Alloy Design of Steels
YOU Wei; LIU Ya-xiu; BAI Bing-zhe; FANG Hong-sheng
2008-01-01
RBF model, a new type of artificial neural network model was developed to design the content of carbon in low-alloy engineering steels. The errors of the ANN model are. MSE 0. 052 1, MSRE 17. 85%, and VOF 1. 932 9. The results obtained are satisfactory. The method is a powerful aid for designing new steels.
Research on Data Mining Based on RBF Neural Network%基于RBF神经网络的数据挖掘研究
徐晓
2014-01-01
This paper discusses the principle and related methods of data mining technology based on data warehouse, and introduces the principle and characteristics of RBF neural network. Aiming at the characteristics of RBF neural network such as strong nonlinear mapping ability and high-speed learning, this paper introduces the data cleaning, pretreatment and regularization steps of data mining method based on RBF neural network. With the distributed information storage feature, the neural network can use a large number of connections between neurons and the analysis of connection weights to limit specific information. The network system built by this idea will not lead to an overall paralysis, even if the local network is damaged..%探讨了基于数据仓库的数据挖掘技术的原理与相关方法，介绍了RBF神经网络的原理与特点。针对RBF神经网络非线性映射能力强和学习速度快等特点，介绍了基于RBF神经网络的数据挖掘方法的数据清洗、预处理和正则化等操作步骤。神经网络具有分布式存储信息的特点，能够利用大量神经元间的连接，以及连接权值的分析，来限定特定信息。使用这种思想构建的网络系统，即使在局部的网络损坏，也不会导致整体的瘫痪。
Hughes, D I; Sikander, S; Kinnon, C M; Boyle, K A; Watanabe, M; Callister, R J; Graham, B A
2012-01-01
Axo-axonic synapses on the central terminals of primary afferent fibres modulate sensory input and are the anatomical correlate of presynaptic inhibition. Although several classes of primary afferents are under such inhibitory control, the origin of these presynaptic inputs in the dorsal horn is unknown. Here, we characterize the neurochemical, anatomical and electrophysiological properties of parvalbumin (PV)-expressing cells in wild-type and transgenic mice where enhanced green fluorescent protein (eGFP) is expressed under the PV promoter. We show that most PV cells have either islet or central cell-like morphology, receive inputs from myelinated primary afferent fibres and are concentrated in laminae II inner and III. We also show that inhibitory PV terminals in lamina II inner selectively target the central terminals of myelinated afferents (∼80% of 935 PVeGFP boutons) and form axo-axonic synapses (∼75% of 71 synapses from PV boutons). Targeted whole-cell patch-clamp recordings from PVeGFP positive cells in laminae II and III showed action potential discharge was restricted to the tonic firing and initial bursting patterns (67% and 33% respectively; n = 18), and virtually all express Ih subthreshold voltage-gated currents (94%; n = 18). These neurons show higher rheobase current than non-eGFP cells but respond with high frequency action potential discharge upon activation. Together, our findings show that PV neurons in laminae II and III are a likely source of inhibitory presynaptic input on to myelinated primary afferents. Consequently PV cells are ideally placed to play an important role in the development of central sensitization and tactile allodynia. PMID:22674718
Hancock, Matthew C; Magnan, Jerry F
2016-10-01
In the assessment of nodules in CT scans of the lungs, a number of image-derived features are diagnostically relevant. Currently, many of these features are defined only qualitatively, so they are difficult to quantify from first principles. Nevertheless, these features (through their qualitative definitions and interpretations thereof) are often quantified via a variety of mathematical methods for the purpose of computer-aided diagnosis (CAD). To determine the potential usefulness of quantified diagnostic image features as inputs to a CAD system, we investigate the predictive capability of statistical learning methods for classifying nodule malignancy. We utilize the Lung Image Database Consortium dataset and only employ the radiologist-assigned diagnostic feature values for the lung nodules therein, as well as our derived estimates of the diameter and volume of the nodules from the radiologists' annotations. We calculate theoretical upper bounds on the classification accuracy that are achievable by an ideal classifier that only uses the radiologist-assigned feature values, and we obtain an accuracy of 85.74 [Formula: see text], which is, on average, 4.43% below the theoretical maximum of 90.17%. The corresponding area-under-the-curve (AUC) score is 0.932 ([Formula: see text]), which increases to 0.949 ([Formula: see text]) when diameter and volume features are included and has an accuracy of 88.08 [Formula: see text]. Our results are comparable to those in the literature that use algorithmically derived image-based features, which supports our hypothesis that lung nodules can be classified as malignant or benign using only quantified, diagnostic image features, and indicates the competitiveness of this approach. We also analyze how the classification accuracy depends on specific features and feature subsets, and we rank the features according to their predictive power, statistically demonstrating the top four to be spiculation, lobulation, subtlety, and
Efficient algorithm for training interpolation RBF networks with equally spaced nodes.
Huan, Hoang Xuan; Hien, Dang Thi Thu; Tue, Huynh Huu
2011-06-01
This brief paper proposes a new algorithm to train interpolation Gaussian radial basis function (RBF) networks in order to solve the problem of interpolating multivariate functions with equally spaced nodes. Based on an efficient two-phase algorithm recently proposed by the authors, Euclidean norm associated to Gaussian RBF is now replaced by a conveniently chosen Mahalanobis norm, that allows for directly computing the width parameters of Gaussian radial basis functions. The weighting parameters are then determined by a simple iterative method. The original two-phase algorithm becomes a one-phase one. Simulation results show that the generality of networks trained by this new algorithm is sensibly improved and the running time significantly reduced, especially when the number of nodes is large.
Liu, Jinkun
2013-01-01
Radial Basis Function (RBF) Neural Network Control for Mechanical Systems is motivated by the need for systematic design approaches to stable adaptive control system design using neural network approximation-based techniques. The main objectives of the book are to introduce the concrete design methods and MATLAB simulation of stable adaptive RBF neural control strategies. In this book, a broad range of implementable neural network control design methods for mechanical systems are presented, such as robot manipulators, inverted pendulums, single link flexible joint robots, motors, etc. Advanced neural network controller design methods and their stability analysis are explored. The book provides readers with the fundamentals of neural network control system design. This book is intended for the researchers in the fields of neural adaptive control, mechanical systems, Matlab simulation, engineering design, robotics and automation. Jinkun Liu is a professor at Beijing University of Aeronautics and Astronauti...
A RBF Based Local Gridfree Scheme for Unsteady Convection-Diffusion Problems
Sanyasiraju VSS Yedida
2009-12-01
Full Text Available In this work a Radial Basis Function (RBF based local gridfree scheme has been presented for unsteady convection diffusion equations. Numerical studies have been made using multiquadric (MQ radial function. Euler and a three stage Runge-Kutta schemes have been used for temporal discretization. The developed scheme is compared with the corresponding finite difference (FD counterpart and found that the solutions obtained using the former are more superior. As expected, for a fixed time step and for large nodal densities, thought the Runge-Kutta scheme is able to maintain higher order of accuracy over the Euler method, the temporal discretization is independent of the improvement in the solution which in the developed scheme has been achived by optimizing the shape parameter of the RBF.
Solving time-dependent problems by an RBF-PS method with an optimal shape parameter
Neves, A M A; Roque, C M C; Ferreira, A J M; Jorge, R M N [Departamento de Engenharia Mecanica e Gestao Industrial, Faculdade de Engenharia da Universidade do Porto, Rua Dr. Roberto Frias, 4200-465 Porto (Portugal); Soares, C M M, E-mail: ana.m.neves@fe.up.p, E-mail: croque@fe.up.p, E-mail: ferreira@fe.up.p, E-mail: cristovao.mota.soares@dem.ist.utl.p, E-mail: rnatal@fe.up.p [IDMEC - Instituto de Engenharia Mecanica - Instituto Superior Tecnico, Av. Rovisco Pais, 1096 Lisboa Codex (Portugal)
2009-08-01
An hybrid technique is used for the solutions of static and time-dependent problems. The idea is to combine the radial basis function (RBF) collocation method and the pseudospectal (PS) method getting to the RBF-PS method. The approach presented in this paper includes a shape parameter optimization and produces highly accurate results. Different examples of the procedure are presented and different radial basis functions are used. One and two-dimensional problems are considered with various boundary and initial conditions. We consider generic problems, but also results on beams and plates. The displacement and the stress analysis are conducted for static and transient dynamic situations. Results obtained are in good agreement with exact solutions or references considered.
A Hybrid RBF-SVM Ensemble Approach for Data Mining Applications
M.Govindarajan
2014-02-01
Full Text Available One of the major developments in machine learning in the past decade is the ensemble method, which finds highly accurate classifier by combining many moderately accurate component classifiers. This paper addresses using an ensemble of classification methods for data mining applications like intrusion detection, direct marketing, and signature verification. In this research work, new hybrid classification method is proposed for heterogeneous ensemble classifiers using arcing and their performances are analyzed in terms of accuracy. A Classifier ensemble is designed using a Radial Basis Function (RBF and Support Vector Machine (SVM as base classifiers. Here, modified training sets are formed by resampling from original training set; classifiers constructed using these training sets and then combined by voting. The proposed RBF-SVM hybrid system is superior to individual approach for intrusion detection, direct marketing, and signature verification in terms of classification accuracy.
Nonlinear Adaptive PID Control for Greenhouse Environment Based on RBF Network
Guanghui Li
2012-04-01
Full Text Available This paper presents a hybrid control strategy, combining Radial Basis Function (RBF network with conventional proportional, integral, and derivative (PID controllers, for the greenhouse climate control. A model of nonlinear conservation laws of enthalpy and matter between numerous system variables affecting the greenhouse climate is formulated. RBF network is used to tune and identify all PID gain parameters online and adaptively. The presented Neuro-PID control scheme is validated through simulations of set-point tracking and disturbance rejection. We compare the proposed adaptive online tuning method with the offline tuning scheme that employs Genetic Algorithm (GA to search the optimal gain parameters. The results show that the proposed strategy has good adaptability, strong robustness and real-time performance while achieving satisfactory control performance for the complex and nonlinear greenhouse climate control system, and it may provide a valuable reference to formulate environmental control strategies for actual application in greenhouse production.
Nonlinear adaptive PID control for greenhouse environment based on RBF network.
Zeng, Songwei; Hu, Haigen; Xu, Lihong; Li, Guanghui
2012-01-01
This paper presents a hybrid control strategy, combining Radial Basis Function (RBF) network with conventional proportional, integral, and derivative (PID) controllers, for the greenhouse climate control. A model of nonlinear conservation laws of enthalpy and matter between numerous system variables affecting the greenhouse climate is formulated. RBF network is used to tune and identify all PID gain parameters online and adaptively. The presented Neuro-PID control scheme is validated through simulations of set-point tracking and disturbance rejection. We compare the proposed adaptive online tuning method with the offline tuning scheme that employs Genetic Algorithm (GA) to search the optimal gain parameters. The results show that the proposed strategy has good adaptability, strong robustness and real-time performance while achieving satisfactory control performance for the complex and nonlinear greenhouse climate control system, and it may provide a valuable reference to formulate environmental control strategies for actual application in greenhouse production.
A RBF neural network model with GARCH errors: Application to electricity price forecasting
Coelho, Leandro dos Santos [Industrial and Systems Engineering Graduate Program, PPGEPS, Pontifical Catholic University of Parana, Imaculada Conceicao, 1155, Zip code 80215-901, Curitiba, Parana (Brazil); Santos, Andre A.P. [Department of Statistics, Universidad Carlos III de Madrid, C/ Madrid, 126, 28903 Getafe, Madrid (Spain)
2011-01-15
In this article, we propose a nonlinear forecasting model based on radial basis function neural networks (RBF-NNs) with Gaussian activation functions and robust clustering algorithms to model the conditional mean and a parametric generalized autoregressive conditional heteroskedasticity (GARCH) specification to model the conditional volatility. Instead of calibrating the parameters of the RBF-NNs via numerical simulations, we propose an estimation procedure by which the number of basis functions, their corresponding widths and the parameters of the GARCH model are jointly estimated via maximum likelihood along with a genetic algorithm to maximize the likelihood function. We use this model to provide multi-step-ahead point and direction-of-change forecasts of the Spanish electricity pool prices. (author)
Adaptive RBF Neural Network Control for Three-Phase Active Power Filter
Juntao Fei
2013-05-01
Full Text Available Abstract An adaptive radial basis function (RBF neural network control system for three-phase active power filter (APF is proposed to eliminate harmonics. Compensation current is generated to track command current so as to eliminate the harmonic current of non-linear load and improve the quality of the power system. The asymptotical stability of the APF system can be guaranteed with the proposed adaptive neural network strategy. The parameters of the neural network can be adaptively updated to achieve the desired tracking task. The simulation results demonstrate good performance, for example showing small current tracking error, reduced total harmonic distortion (THD, improved accuracy and strong robustness in the presence of parameters variation and nonlinear load. It is shown that the adaptive RBF neural network control system for three-phase APF gives better control than hysteresis control.
Research on Space Vector PWM Based on RBF Neural Network%基于RBF神经网络的SVPWM研究
宣光银; 胡丹; 车畅
2011-01-01
论述了电压空间矢量脉宽调制( SVPWM)的基本原理,提出利用径向基(RBF)网络实现SVPWM的方法；并在Matlab下结合神经网络工具箱进了仿真研究.仿真结果表明,RBF-SVPWM收敛速度快、误差精度高,使用该方法建立的异步电动机控制系统具有更小的定子电流谐波和转矩脉动.%The basic principle of space voltage vector PWM was discussed,and a new algorithm for SVPWM based on radial basis function neural network( RBF-SVPWM) was proposed. The emulation of RBF-SVPWM was carried out in Mat-lab together with Nntoolbox. The results indicate that RBF-SVPWM has a fast convergence and high accurate error,and the induction motor control system using RBF-SVPWM also has less stator current harmonics and torque pulsation.
2D Sketch based recognition of 3D freeform shapes by using the RBF Neural Network
Qin, S F; Sun, Guangmin; Wright, D K; Lim, S.; Khan, U.; Mao, C.
2005-01-01
This paper presents a novel free-form surface recognition method from 2D freehand sketching. The approach is based on the Radial basis function (RBF), an artificial intelligence technique. A simple three-layered network has been designed and constructed. After training and testing with two types of surfaces (four sided boundary surfaces and four close section surfaces), it has been shown that the method is useful in freeform surface recognition. The testing results are very satisfactory.
2D sketch based recognition of 3D freeform shape by using the RBF neural network
Qin, SF; Sun, GM; Wright, DK; Lim, S.; Khan, U.; Mao, C.
2005-01-01
This paper presents a novel free-form surface recognition method from 2D freehand sketching. The approach is based on the Radial basis function (RBF), an artificial intelligence technique. A simple three-layered network has been designed and constructed. After training and testing with two types of surfaces (four sided boundary surfaces and four close section surfaces), it has been shown that the method is useful in freeform surface recognition. The testing results are very satisfactory.
张军朝; 陈俊杰
2011-01-01
Photovoltaic Array is nonlinear, and the power generated by it is influenced by sun light, temperature and so on. We put forward PV array model by using neural networks identification technique in this paper. The temperature, radiation and voltage of the solar cells are taken as the input and the current as the output of the neural networks model. Using RBF neural network to model for photovoltaic battery and particle swarm optimization algorithm to optimize the RBF neural network, finally the photovoltaic model is established. Simulated experiments are carried out on the photovoltaic battery data, the results show that the improved RBF neural networks have better accuracy and adapt ability than traditional RBF method. The RBF neural networks modeling makes it possible to design on-line controller of photovoltaic system.%本文研究神经网络在光伏电池建模优化问题.由于光伏电池具有高度非线性特性,其输出功率受到外界自然因素的影响,使得传统方法不能满足光伏控制系统动态要求.针对上述问题,本文提出一种粒子群优化的神经网络光伏电池建模算法.改进的方法以日照、温度和负载电压作为提出的RBF神经网络模型的输入值,把光伏电池的输出功率作为神经网络的输出,采用RBF神经网络对光伏电池进行建模,同时利用粒子群算法对神经网络参数进行优化,最后建立光伏电池的动态响应模型.仿真实验结果证明,所提模型更好地克服传统方法的缺点,收敛速度快,具有较高的预测精度和适合能力.
Design of SVPWM Overmodulation Based on RBF Neural Networks%基于RBF神经网络的SVPWM过调制
易灵芝; 何素芬; 李举成; 彭寒梅; 李明; 陈彦如
2009-01-01
线性调制状态下的逆变器,存在不能充分利用直流母线电压的问题,为了获得尽可能大的输出电压,一般对逆变器进行过调制控制.由于过调制的非线性,计算复杂,提出新的基于神经网络的SVPWM逆变器,采用线性调制和过调制2种模式,通过限定轨迹双调制模式法,实现在整个调制范围内线性控制.采用资源分配法确定径向基函数的网络结构和参数,设计出较精简的网络实现SVPWM;并将这种逆变器应用于异步电动机控制系统中.最后在Matlab环境下建立基于神经网络的SVPWM逆变器供电的异步电机控制系统仿真模型, 仿真结果表明,该方法简单、高效、控制效果良好,能提高直流母线电压利用率,降低输出电流谐波含量和电机转矩脉动.%The problem of obtaining the higher output voltage value in linear modulation condition PWM inversion is discussed.A complete artificial-neural-network space vector pulse width modulation(ANN-SVM)controller for a voltage source inverter is presented.The operation is very well both in under-modulation and over-modulation regions.The linear control is realized in the entire modulation scope by limit trajectories.The resources method of distribution determination RBF network architecture and the parameter are used.The RBF neural network realize the voltage space vector modulate to control the inverter,which is applied to the control system of asynchronous machine.The simulation results show that the method features simplicity,high efficiency and excellent control effect.The use ratio of DC bus voltage is raised,and the output current harmonics content and the torque pulsation are reduced.
Skretas, Sotirios B.; Papadopoulos, Demetrios P. [Electrical Machines Laboratory, Department of Electrical and Computer Engineering, Democritos University of Thrace (DUTH), 12 V. Sofias, 67100 Xanthi (Greece)
2009-09-15
In this paper an efficient design along with modeling and simulation of a transformer-less small-scale centralized DC - bus Grid Connected Hybrid (Wind-PV) power system for supplying electric power to a single phase of a three phase low voltage (LV) strong distribution grid are proposed and presented. The main components of the hybrid system are: a PV generator (PVG); and an array of horizontal-axis, fixed-pitch, small-size, variable-speed wind turbines (WTs) with direct-driven permanent magnet synchronous generator (PMSG) having an embedded uncontrolled bridge rectifier. An overview of the basic theory of such systems along with their modeling and simulation via Simulink/MATLAB software package are presented. An intelligent control method is applied to the proposed configuration to simultaneously achieve three desired goals: to extract maximum power from each hybrid power system component (PVG and WTs); to guarantee DC voltage regulation/stabilization at the input of the inverter; to transfer the total produced electric power to the electric grid, while fulfilling all necessary interconnection requirements. Finally, a practical case study is conducted for the purpose of fully evaluating a possible installation in a city site of Xanthi/Greece, and the practical results of the simulations are presented. (author)
Lam, Dao; Wunsch, Donald
2017-01-01
Ever-increasing size and complexity of data sets create challenges and potential tradeoffs of accuracy and speed in learning algorithms. This paper offers progress on both fronts. It presents a mechanism to train the unsupervised learning features learned from only one layer to improve performance in both speed and accuracy. The features are learned by an unsupervised feature learning (UFL) algorithm. Then, those features are trained by a fast radial basis function (RBF) extreme learning machine (ELM). By exploiting the massive parallel computing attribute of modern graphics processing unit, a customized compute unified device architecture (CUDA) kernel is developed to further speed up the computing of the RBF kernel in the ELM. Results tested on Canadian Institute for Advanced Research and Mixed National Institute of Standards and Technology data sets confirm the UFL RBF ELM achieves high accuracy, and the CUDA implementation is up to 20 times faster than CPU and the naive parallel approach.
薛晓岑; 向文国; 吕剑虹
2014-01-01
针对热工过程的非线性辨识问题，提出了一种基于差分进化算法（ DE ）的径向基函数神经网络（ RBFNN）模型设计方法。该方法将DE算法的种群分解为几组并行的子种群，每组子种群对应于一类隐节点数相同的RBF网络。在RBFNN的学习过程中进行多子种群并行优化，从而实现RBF网络结构与参数的同时调整。算法可以利用热工对象的输入输出数据，自动设计出满足误差精度要求且结构较小的RBFNN模型。然后将该算法应用于热工对象的辨识，对于单输入单输出系统，得到的RBFNN模型只需1个隐节点。对于多输入单输出系统，RBF网络也仅需较少的隐层节点。仿真结果表明，用该方法设计的RBFNN模型结构简单，且辨识误差小，具有较好的泛化能力。%For the nonlinear identification of thermal process, a new radial basis function neural net-work ( RBFNN) design method is proposed based on the differential evolution algorithm ( DE) .In the method, the population in the DE algorithm is divided into several parallel subpopulations, and each subpopulation corresponds to a class of RBF network solutions with the same hidden nodes.In the RBFNN learning process, the network structure and parameters are adjusted simultaneously through the parallel optimization of the subpopulations.Under the given error limit, the algorithm can design an RBF model automatically with fewer hidden nodes according to thermal input and out-put data.Then, the algorithm is used to identify nonlinear thermal processes.For single-input sin-gle-output system identification, only one node is required in the RBFNN hidden layer.For multi-in-put single-output system identification, the RBFNN model also requires less hidden nodes.The sim-ulation results show that the proposed approach can achieve the given identification accuracy with fe-wer hidden nodes, and has good generalization ability.
熊亮; 赵俊锴
2015-01-01
Assessment of compressive strength of substation concrete column is an important foundation of damage degree and bearing capacity of construction.An RBF neural network model (RBF-NN )is applied to assessing compressive strength of concrete by ultrasonic and rebound combined method.An experimental method is given for compressive strength of concrete test by ultrasonic and rebound combined method.It is proved that RBF-NN model has higher evaluation precision than that of regression calculation by experimental test and emulation analysis.%变电站混凝土立柱抗压强度的评定是判断变电站混凝土结构损伤程度、剩余承载力的重要依据。设计了一个 RBF 神经网络模型，将其应用于超声回弹综合法评定变电站混凝土立柱抗压强度，给出了用超声回弹法进行混凝土强度测试的方法。经试验测试和仿真分析表明，所提出的 RBF 神经网络比传统的回归计算方法具有更高的评估精度。
Application of RBF Neural Network in Optimizing Machining Parameters%径向基函数网络在优化机械加工参数中的应用
朱喜林; 吴博达; 武星星
2004-01-01
In machining processes, errors of rough in dimension, shape and location lead to changes in processing quantity, and the material of a workpiece may not be uniform. For these reasons, cutting force changes in machining, making the machining system deformable. Consequently errors in workpieces may occur. This is called the error reflection phenomenon. Generally, such errors can be reduced through repeated processing while using appropriate processing quantity in each processing based on operator's experience.According to the theory of error reflection, the error reflection coefficient indicates the extent to which errors of rough influence errors of workpieces. It is related to several factors such as machining condition, hardness of the workpiece, etc. This non-linear relation cannot be worked out using any formula. RBF neural network can approximate a non-linear function within any precision and be trained fast. In this paper, non-linear mapping ability of a fuzzy-neural network is utilized to approximate the non-linear relation. After training of the network with swatch collection obtained in experiments, an appropriate output can be obtained when an input is given. In this way, one can get the required number of processing and the processing quantity each time from the machining condition. Angular rigidity of a machining system,hardness of workpiece, etc., can be input in a form of fuzzy values. Feasibility in solving error reflection and optimizing machining parameters with a RBF neural network is verified by a simulation test with MATLAB.
Application of Nonlinear Predictive Control Based on RBF Network Predictive Model in MCFC Plant
CHEN Yue-hua; CAO Guang-yi; ZHU Xin-jian
2007-01-01
This paper described a nonlinear model predictive controller for regulating a molten carbonate fuel cell (MCFC). A detailed mechanism model of output voltage of a MCFC was presented at first. However, this model was too complicated to be used in a control system. Consequently, an off line radial basis function (RBF) network was introduced to build a nonlinear predictive model. And then, the optimal control sequences were obtained by applying golden mean method. The models and controller have been realized in the MATLAB environment. Simulation results indicate the proposed algorithm exhibits satisfying control effect even when the current densities vary largely.
Lukas Falat
2014-01-01
Full Text Available In this paper, authors apply feed-forward artificial neural network (ANN of RBF type into the process of modelling and forecasting the future value of USD/CAD time series. Authors test the customized version of the RBF and add the evolutionary approach into it. They also combine the standard algorithm for adapting weights in neural network with an unsupervised clustering algorithm called K-means. Finally, authors suggest the new hybrid model as a combination of a standard ANN and a moving average for error modeling that is used to enhance the outputs of the network using the error part of the original RBF. Using high-frequency data, they examine the ability to forecast exchange rate values for the horizon of one day. To determine the forecasting efficiency, authors perform the comparative out-of-sample analysis of the suggested hybrid model with statistical models and the standard neural network.
Numerical Solution of Stokes Flow in a Circular Cavity Using Mesh-free Local RBF-DQ
Kutanaai, S Soleimani; Roshan, Naeem; Vosoughi, A;
2012-01-01
This work reports the results of a numerical investigation of Stokes flow problem in a circular cavity as an irregular geometry using mesh-free local radial basis function-based differential quadrature (RBF-DQ) method. This method is the combination of differential quadrature approximation...... is applied on a two-dimensional geometry. The obtained results from the numerical simulations are compared with those gained by previous works. Outcomes prove that the current technique is in very good agreement with previous investigations and this fact that RBF-DQ method is an accurate and flexible method...... in solution of partial differential equations (PDEs)....
Lei Wang
2014-01-01
Full Text Available Offshore floating wind turbine (OFWT has been a challenging research spot because of the high-quality wind power and complex load environment. This paper focuses on the research of variable torque control of offshore wind turbine on Spar floating platform. The control objective in below-rated wind speed region is to optimize the output power by tracking the optimal tip-speed ratio and ideal power curve. Aiming at the external disturbances and nonlinear uncertain dynamic systems of OFWT because of the proximity to load centers and strong wave coupling, this paper proposes an advanced radial basis function (RBF neural network approach for torque control of OFWT system at speeds lower than rated wind speed. The robust RBF neural network weight adaptive rules are acquired based on the Lyapunov stability analysis. The proposed control approach is tested and compared with the NREL baseline controller using the “NREL offshore 5 MW wind turbine” model mounted on a Spar floating platform run on FAST and Matlab/Simulink, operating in the below-rated wind speed condition. The simulation results show a better performance in tracking the optimal output power curve, therefore, completing the maximum wind energy utilization.
Tatar Afshin
2016-03-01
Full Text Available Raw natural gases usually contain water. It is very important to remove the water from these gases through dehydration processes due to economic reasons and safety considerations. One of the most important methods for water removal from these gases is using dehydration units which use Triethylene glycol (TEG. The TEG concentration at which all water is removed and dew point characteristics of mixture are two important parameters, which should be taken into account in TEG dehydration system. Hence, developing a reliable and accurate model to predict the performance of such a system seems to be very important in gas engineering operations. This study highlights the use of intelligent modeling techniques such as Multilayer perceptron (MLP and Radial Basis Function Neural Network (RBF-ANN to predict the equilibrium water dew point in a stream of natural gas based on the TEG concentration of stream and contractor temperature. Literature data set used in this study covers temperatures from 10 °C to 80 °C and TEG concentrations from 90.000% to 99.999%. Results showed that both models are accurate in prediction of experimental data and the MLP model gives more accurate predictions compared to RBF model.
Ishii-Minami, Naoko; Kawahara, Yoshihiro; Yoshida, Yuri; Okada, Kazunori; Ando, Sugihiro; Matsumura, Hideo; Terauchi, Ryohei; Minami, Eiichi; Nishizawa, Yoko
2016-01-01
Magnaporthe oryzae, the fungus causing rice blast disease, should contend with host innate immunity to develop invasive hyphae (IH) within living host cells. However, molecular strategies to establish the biotrophic interactions are largely unknown. Here, we report the biological function of a M. oryzae-specific gene, Required-for-Focal-BIC-Formation 1 (RBF1). RBF1 expression was induced in appressoria and IH only when the fungus was inoculated to living plant tissues. Long-term successive imaging of live cell fluorescence revealed that the expression of RBF1 was upregulated each time the fungus crossed a host cell wall. Like other symplastic effector proteins of the rice blast fungus, Rbf1 accumulated in the biotrophic interfacial complex (BIC) and was translocated into the rice cytoplasm. RBF1-knockout mutants (Δrbf1) were severely deficient in their virulence to rice leaves, but were capable of proliferating in abscisic acid-treated or salicylic acid-deficient rice plants. In rice leaves, Δrbf1 inoculation caused necrosis and induced defense-related gene expression, which led to a higher level of diterpenoid phytoalexin accumulation than the wild-type fungus did. Δrbf1 showed unusual differentiation of IH and dispersal of the normally BIC-focused effectors around the short primary hypha and the first bulbous cell. In the Δrbf1-invaded cells, symplastic effectors were still translocated into rice cells but with a lower efficiency. These data indicate that RBF1 is a virulence gene essential for the focal BIC formation, which is critical for the rice blast fungus to suppress host immune responses. PMID:27711180
V. Bayona
2015-04-01
Full Text Available A numerical model based on Radial Basis Function-generated Finite Differences (RBF-FD is developed for simulating the Global Electric Circuit (GEC within the Earth's atmosphere, represented by a 3-D variable coefficient linear elliptic PDE in a spherically-shaped volume with the lower boundary being the Earth's topography and the upper boundary a sphere at 60 km. To our knowledge, this is (1 the first numerical model of the GEC to combine the Earth's topography with directly approximating the differential operators in 3-D space, and related to this (2 the first RBF-FD method to use irregular 3-D stencils for discretization to handle the topography. It benefits from the mesh-free nature of RBF-FD, which is especially suitable for modeling high-dimensional problems with irregular boundaries. The RBF-FD elliptic solver proposed here makes no limiting assumptions on the spatial variability of the coefficients in the PDE (i.e. the conductivity profile, the right hand side forcing term of the PDE (i.e. distribution of current sources or the geometry of the lower boundary.
Joseph Ahlander
Full Text Available BACKGROUND: The retinoblastoma (Rb tumor suppressor protein can function as a DNA replication inhibitor as well as a transcription factor. Regulation of DNA replication may occur through interaction of Rb with the origin recognition complex (ORC. PRINCIPAL FINDINGS: We characterized the interaction of Drosophila Rb, Rbf1, with ORC. Using expression of proteins in Drosophila S2 cells, we found that an N-terminal Rbf1 fragment (amino acids 1-345 is sufficient for Rbf1 association with ORC but does not bind to dE2F1. We also found that the C-terminal half of Rbf1 (amino acids 345-845 interacts with ORC. We observed that the amino-terminal domain of Rbf1 localizes to chromatin in vivo and associates with chromosomal regions implicated in replication initiation, including colocalization with Orc2 and acetylated histone H4. CONCLUSIONS/SIGNIFICANCE: Our results suggest that Rbf1 can associate with ORC and chromatin through domains independent of the E2F binding site. We infer that Rbf1 may play a role in regulating replication directly through its association with ORC and/or chromatin factors other than E2F. Our data suggest an important role for retinoblastoma family proteins in cell proliferation and tumor suppression through interaction with the replication initiation machinery.
曹龙汉; 刘小丽; 郭晓东; 王申涛; 代睿
2011-01-01
针对气门故障,以缸盖振动信号的小波包能量谱作为故障特征参数,提出一种粗糙集(RS)与改进的量了微粒群径向基函数神经网络(QPSO-RBF NN)相结合的故障诊断方法.首先应用粗糙集对试验所得的特征参数进行属性约简,去掉冗余信息,简化RBF网络的结构；然后将带变异算子的QPSO算法引入到RBF网络的学习过程中,改进其现有的学习算法,进一步提高故障预测能力.通过对6135D型柴油机气门故障进行诊断,结果表明该方法提高了诊断的精度和效率.%A fault diagnosis method of combining RS (rough set) and improved QPSO-RBF NN (quantum-behaved particle swarm optimization-radial basis function neural network) is proposed for valve fault, in which wavelet packet energy spectrum from vibration signals of bonnet is taken as fault characteristic parameters. Firstly, attributes of characteristic parameters are reduced by RS theory in order to delete redundant attributes and simplify the inputs of RBF NN, then QPSO algorithm with mutation operator is introduced into the learning process of RBF NN to improve its existing learning algorithms and enhance its predictive ability. The simulation results of valve fault on 6135D type diesel engine show that the method enhancs accuracy and efficiency of fault diagnosis.
Lihua Yang
2015-04-01
Full Text Available In order to improve the accuracy of grain production forecasting, this study proposed a new combination forecasting model, the model combined stepwise regression method with RBF neural network by assigning proper weights using inverse variance method. By comparing different criteria, the result indicates that the combination forecasting model is superior to other models. The performance of the models is measured using three types of error measurement, which are Mean Absolute Percentage Error (MAPE, Theil Inequality Coefficient (Theil IC and Root Mean Squared Error (RMSE. The model with smallest value of MAPE, Theil IC and RMSE stands out to be the best model in predicting the grain production. Based on the MAPE, Theil IC and RMSE evaluation criteria, the combination model can reduce the forecasting error and has high prediction accuracy in grain production forecasting, making the decision more scientific and rational.
Adaptive Global Sliding Mode Control for MEMS Gyroscope Using RBF Neural Network
Yundi Chu
2015-01-01
Full Text Available An adaptive global sliding mode control (AGSMC using RBF neural network (RBFNN is proposed for the system identification and tracking control of micro-electro-mechanical system (MEMS gyroscope. Firstly, a new kind of adaptive identification method based on the global sliding mode controller is designed to update and estimate angular velocity and other system parameters of MEMS gyroscope online. Moreover, the output of adaptive neural network control is used to adjust the switch gain of sliding mode control dynamically to approach the upper bound of unknown disturbances. In this way, the switch item of sliding mode control can be converted to the output of continuous neural network which can weaken the chattering in the sliding mode control in contrast to the conventional fixed gain sliding mode control. Simulation results show that the designed control system can get satisfactory tracking performance and effective estimation of unknown parameters of MEMS gyroscope.
Nuclear power plant fault diagnosis based on genetic-RBF neural network
SHI Xiao-cheng; XIE Chun-ling; WANG Yuan-hui
2006-01-01
It is necessary to develop an automatic fault diagnosis system to avoid a possible nuclear disaster caused by an inaccurate fault diagnosis in the nuclear power plant by the operator. Because Radial Basis Function Neural Network (RBFNN) has the characteristics of optimal approximation and global approximation. The mixed coding of binary system and decimal system is introduced to the structure and parameters of RBFNN, which is trained in course of the genetic optimization. Finally, a fault diagnosis system according to the frequent faults in condensation and feed water system of nuclear power plant is set up. As a result, Genetic-RBF Neural Network (GRBFNN) makes the neural network smaller in size and higher in generalization ability. The diagnosis speed and accuracy are also improved.
ZHANG Yongzhi
2016-10-01
Full Text Available A dynamic fuzzy RBF neural network model was built to predict the mechanical properties of welded joints, and the purpose of the model was to overcome the shortcomings of static neural networks including structural identification, dynamic sample training and learning algorithm. The structure and parameters of the model are no longer head of default, dynamic adaptive adjustment in the training, suitable for dynamic sample data for learning, learning algorithm introduces hierarchical learning and fuzzy rule pruning strategy, to accelerate the training speed of model and make the model more compact. Simulation of the model was carried out by using three kinds of thickness and different process TC4 titanium alloy TIG welding test data. The results show that the model has higher prediction accuracy, which is suitable for predicting the mechanical properties of welded joints, and has opened up a new way for the on-line control of the welding process.
GONG Huanchun
2014-01-01
In order to diagnose the unit economic performance online,the radial basis function (RBF) process neural network with two hidden layers was introduced to online prediction of steam turbine exhaust enthalpy.Thus,the model reflecting complicated relationship between the steam turbine exhaust enthalpy and the relative operation parameters was established.Moreover,the enthalpy of final stage extraction steam and exhaust from a 300 MW unit turbine was taken as the example to perform the online calculation. The results show that,the average relative error of this method is less than 1%,so the accuracy of this al-gorithm is higher than that of the BP neutral network.Furthermore,this method has advantages of high convergence rate,simple structure and high accuracy.
An Adaptive-PSO-Based Self-Organizing RBF Neural Network.
Han, Hong-Gui; Lu, Wei; Hou, Ying; Qiao, Jun-Fei
2016-10-24
In this paper, a self-organizing radial basis function (SORBF) neural network is designed to improve both accuracy and parsimony with the aid of adaptive particle swarm optimization (APSO). In the proposed APSO algorithm, to avoid being trapped into local optimal values, a nonlinear regressive function is developed to adjust the inertia weight. Furthermore, the APSO algorithm can optimize both the network size and the parameters of an RBF neural network simultaneously. As a result, the proposed APSO-SORBF neural network can effectively generate a network model with a compact structure and high accuracy. Moreover, the analysis of convergence is given to guarantee the successful application of the APSO-SORBF neural network. Finally, multiple numerical examples are presented to illustrate the effectiveness of the proposed APSO-SORBF neural network. The results demonstrate that the proposed method is more competitive in solving nonlinear problems than some other existing SORBF neural networks.
基于PF-RBF神经网络的短期风电功率预测%Short-Term Wind Power Prediction Based on PF-RBF Neural Network
王永翔; 陈国初; 张鑫
2014-01-01
为了提高风电功率的预测精度，研究了一种基于粒子滤波(PF)与径向基函数(RBF)神经网络相结合的风电功率预测方法。使用PF算法对历史风速数据进行滤波处理，将处理后的风速数据结合风向、温度的历史数据，归一化后构成风电功率预测模型的新的输入数据；利用处理后的新的输入数据和输出数据，建立PF-RBF神经网络预测模型，预测风电场的输出功率。仿真结果表明，使用该预测模型进行风电功率预测，预测精度有一定的提高，连续120 h功率预测的平均绝对百分误差达到8.04%，均方根误差达到10.67%。%To improve accuracy of wind power prediction,this paper proposes a short-term wind power prediction method combining a particle filter (PF)and a radial basis function (RBF) neural network.Historical wind speed data are first processed with a particle filter.The processed wind speed data combined with the historical data of wind direction and temperature are used as in-put to the model.A PF-RBF neural network of wind power output prediction model is established using the new input data.Simulation results show that the proposed model is accurate in wind power prediction.The mean absolute percentage prediction error in a period of 120 hours has been reduced to 8.04%,and the root mean square error is 10.67%.
Şen, Baha; Peker, Musa; Çavuşoğlu, Abdullah; Çelebi, Fatih V
2014-03-01
Sleep scoring is one of the most important diagnostic methods in psychiatry and neurology. Sleep staging is a time consuming and difficult task undertaken by sleep experts. This study aims to identify a method which would classify sleep stages automatically and with a high degree of accuracy and, in this manner, will assist sleep experts. This study consists of three stages: feature extraction, feature selection from EEG signals, and classification of these signals. In the feature extraction stage, it is used 20 attribute algorithms in four categories. 41 feature parameters were obtained from these algorithms. Feature selection is important in the elimination of irrelevant and redundant features and in this manner prediction accuracy is improved and computational overhead in classification is reduced. Effective feature selection algorithms such as minimum redundancy maximum relevance (mRMR); fast correlation based feature selection (FCBF); ReliefF; t-test; and Fisher score algorithms are preferred at the feature selection stage in selecting a set of features which best represent EEG signals. The features obtained are used as input parameters for the classification algorithms. At the classification stage, five different classification algorithms (random forest (RF); feed-forward neural network (FFNN); decision tree (DT); support vector machine (SVM); and radial basis function neural network (RBF)) classify the problem. The results, obtained from different classification algorithms, are provided so that a comparison can be made between computation times and accuracy rates. Finally, it is obtained 97.03 % classification accuracy using the proposed method. The results show that the proposed method indicate the ability to design a new intelligent assistance sleep scoring system.
高丙坤; 郑仁谦; 尹淑欣; 张莉; 岳武峰
2016-01-01
为了正确判断管道是否发生泄漏，本文采用混合学习方法对网络进行训练学习。通过将管道运行参数作为神经网络的输入，管道运行状态作为神经网络的输出，实现两者的非线性映射，以此来判断输入信号是否为泄漏信号，并选用K-means聚类方法和递推最小二乘法来确定网络参数。通过用天然气管道运行的实测数据对RBF神经网络进行了训练和测试，得到结果误差在可接受的范围内，从而证明RBF神经网络的方法可用于天然气管道泄漏检测的研究。%In order to correctly determine whether pipeline leakage occurs, this paper adopts a hybrid learning method for network training. We set the pipeline operation parameters as the input of neural network and running status of the pipe as the neural network output, realizing the two nonlinear mapping, in order to determine whether the input signal is leakage signal , and select K-means clustering method and the recursive least square method to determine the network parameters. With the measurements of the gas pipeline operation on training and testing the RBF neural network, we get the results in an acceptable error range, which prove that the method of RBF neural network can be used for natural gas pipeline leak detection.
Michelle S Longworth
Full Text Available Previously, we discovered a conserved interaction between RB proteins and the Condensin II protein CAP-D3 that is important for ensuring uniform chromatin condensation during mitotic prophase. The Drosophila melanogaster homologs RBF1 and dCAP-D3 co-localize on non-dividing polytene chromatin, suggesting the existence of a shared, non-mitotic role for these two proteins. Here, we show that the absence of RBF1 and dCAP-D3 alters the expression of many of the same genes in larvae and adult flies. Strikingly, most of the genes affected by the loss of RBF1 and dCAP-D3 are not classic cell cycle genes but are developmentally regulated genes with tissue-specific functions and these genes tend to be located in gene clusters. Our data reveal that RBF1 and dCAP-D3 are needed in fat body cells to activate transcription of clusters of antimicrobial peptide (AMP genes. AMPs are important for innate immunity, and loss of either dCAP-D3 or RBF1 regulation results in a decrease in the ability to clear bacteria. Interestingly, in the adult fat body, RBF1 and dCAP-D3 bind to regions flanking an AMP gene cluster both prior to and following bacterial infection. These results describe a novel, non-mitotic role for the RBF1 and dCAP-D3 proteins in activation of the Drosophila immune system and suggest dCAP-D3 has an important role at specific subsets of RBF1-dependent genes.
Kai Heimel
Full Text Available In the phytopathogenic basidiomycete Ustilago maydis, sexual and pathogenic development are tightly connected and controlled by the heterodimeric bE/bW transcription factor complex encoded by the b-mating type locus. The formation of the active bE/bW heterodimer leads to the formation of filaments, induces a G2 cell cycle arrest, and triggers pathogenicity. Here, we identify a set of 345 bE/bW responsive genes which show altered expression during these developmental changes; several of these genes are associated with cell cycle coordination, morphogenesis and pathogenicity. 90% of the genes that show altered expression upon bE/bW-activation require the zinc finger transcription factor Rbf1, one of the few factors directly regulated by the bE/bW heterodimer. Rbf1 is a novel master regulator in a multilayered network of transcription factors that facilitates the complex regulatory traits of sexual and pathogenic development.
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.
Heimel, Kai; Scherer, Mario; Vranes, Miroslav; Wahl, Ramon; Pothiratana, Chetsada; Schuler, David; Vincon, Volker; Finkernagel, Florian; Flor-Parra, Ignacio; Kämper, Jörg
2010-08-05
In the phytopathogenic basidiomycete Ustilago maydis, sexual and pathogenic development are tightly connected and controlled by the heterodimeric bE/bW transcription factor complex encoded by the b-mating type locus. The formation of the active bE/bW heterodimer leads to the formation of filaments, induces a G2 cell cycle arrest, and triggers pathogenicity. Here, we identify a set of 345 bE/bW responsive genes which show altered expression during these developmental changes; several of these genes are associated with cell cycle coordination, morphogenesis and pathogenicity. 90% of the genes that show altered expression upon bE/bW-activation require the zinc finger transcription factor Rbf1, one of the few factors directly regulated by the bE/bW heterodimer. Rbf1 is a novel master regulator in a multilayered network of transcription factors that facilitates the complex regulatory traits of sexual and pathogenic development.
Wang, Zhongqi; Yang, Bo; Kang, Yonggang; Yang, Yuan
2016-01-01
Fixture plays an important part in constraining excessive sheet metal part deformation at machining, assembly, and measuring stages during the whole manufacturing process. However, it is still a difficult and nontrivial task to design and optimize sheet metal fixture locating layout at present because there is always no direct and explicit expression describing sheet metal fixture locating layout and responding deformation. To that end, an RBF neural network prediction model is proposed in this paper to assist design and optimization of sheet metal fixture locating layout. The RBF neural network model is constructed by training data set selected by uniform sampling and finite element simulation analysis. Finally, a case study is conducted to verify the proposed method.
姚应水; 叶明全
2011-01-01
目的 RBF神经网络是一种重要的数据挖掘分类模型,探讨RBF神经网络在解决判别分析问题中的应用.方法 通过实例比较RBF神经网络和logistic回归模型的性能优劣.结果 RBF神经网络的回代拟合效果和泛化能力明显优于logistic回归模型.结论RBF神经网络在医学统计学领域中具有较好的应用前景.%Objective RBF neural network is an important data mining classification model in data mining. To explore the application of RBF neural network on medical discriminant analysis through comparing with logistic regression. Methods Comparing the prediction results by some statistical indexes of the RBF neural network and the logistic regression by using an example. Results The comparison results of the prediction performance between RBF neural network and logistic regression show that RBF neural network is much better than logistic regression for the data. Conclusion RBF neural network will make a better facture of its appfi-cadon in medical researches.
Aoki, Y; Ishii, N; Watanabe, M; Yoshihara, F; Arisawa, M
1998-01-01
The major fungal pathogen for fungal diseases which have become a major medical problem in the last few years is Candida albicans, which can grow both in yeast and hyphae forms. This ability of C. albicans is thought to contribute to its colonization and dissemination within host tissues. In a recent few years, accompanying the introduction of molecular biological tools into C. albicans organism, several factors involved in the signal transduction pathway for yeast-hyphal transition have been identified. One MAP kinase pathway in C. albicans, similar to that leading to STE12 activation in Saccharomyces cerevisiae, has been reported. C. albicans strains mutant in these genes show retarded filamentous growth on a solid media but no impairment of filamentous growth in mice. These results suggest two scenarios that a kinase signaling cascade plays a part in stimulating the morphological transition in C. albicans, and that there would be another signaling pathway effective in animals. In this latter true hyphal pathway, although some candidate proteins, such as Efg1 (transcription factor), Int1 (integrin-like membrane protein), or Phr1 (pH-regulated membrane protein), have been identified, it is still too early to say that we understand the whole picture of that cascade. We have cloned a C. albicans gene encoding a novel DNA binding protein, Rbf1, that predominantly localizes in the nucleus, and shows transcriptional activation capability. Disruption of the functional RBF1 genes of C. albicans induced the filamentous growth on all solid and liquid media tested, suggesting that Rbf1 might be another candidate for the true hyphal pathway. Relationships with other factors described above, and the target (regulated) genes of Rbf1 is under investigation.
Gao, Xiangdong; Chen, Yuquan; You, Deyong; Xiao, Zhenlin; Chen, Xiaohui
2017-02-01
An approach for seam tracking of micro gap weld whose width is less than 0.1 mm based on magneto optical (MO) imaging technique during butt-joint laser welding of steel plates is investigated. Kalman filtering(KF) technology with radial basis function(RBF) neural network for weld detection by an MO sensor was applied to track the weld center position. Because the laser welding system process noises and the MO sensor measurement noises were colored noises, the estimation accuracy of traditional KF for seam tracking was degraded by the system model with extreme nonlinearities and could not be solved by the linear state-space model. Also, the statistics characteristics of noises could not be accurately obtained in actual welding. Thus, a RBF neural network was applied to the KF technique to compensate for the weld tracking errors. The neural network can restrain divergence filter and improve the system robustness. In comparison of traditional KF algorithm, the RBF with KF was not only more effectively in improving the weld tracking accuracy but also reduced noise disturbance. Experimental results showed that magneto optical imaging technique could be applied to detect micro gap weld accurately, which provides a novel approach for micro gap seam tracking.
崔维; 丁玲
2016-01-01
为了提高采摘机器人自主导航和路径规划能力，提出了基于计算机视觉路径规划和 RBF 神经网络自适应逼近算法的导航方法。使用图像分割、平滑处理和边缘检测技术，根据图像像素灰度值确定了导航线的位置，利用逐行扫描的方法得到了导航离散点。路径规划和跟踪使用RBF神经网络逼近算法，通过逼近误差和权值控制路径跟踪的精度，系统响应的执行端使用液压伺服系统，提高了机器人自主导航的精度。以黄瓜采摘作为研究对象，在日光温室对机器人采摘作业进行了测试，通过测试得到了 RBF 神经网络的路径跟踪误差曲线。测试结果表明：机器人可以很好地逼近跟踪规划路径，其计算精度较高，跟踪效果较好。%In order to improve the ability of autonomous navigation and path planning of picking robot , a navigation meth-od is proposed based on computer vision path planning and RBF neural network adaptive approximation algorithm .The use of image segmentation , smoothing and edge detection technology ,the navigation line positions are determined accord-ing to the image pixel gray value using progressive scan method of navigation discrete points .The path planning and tracking using RBF neural network approximation algorithm , the accuracy of the system response is controlled by the ac-curacy of the error and weight control .Taking cucumber as the research object , it tested the robot picking operation in greenhouse , and obtained the path tracking error curve of RBF neural network .The test results show that the robot can get a good approximation of the path .
Curvature estimation of Point-Sampled model based on RBF%基于RBF的点模型曲率估算
张利军; 张若男
2016-01-01
Curvature estimation of point model is the basic work of the point geometry processing , which plays an important role in the follow-up works, such as point clouds simplification and feature extraction, etc. Based on radial basis functions (RBF), an algorithm is presented for effectively estimating curvatures on point-sampled model in this paper. Firstly, the nearest neighbors of each sampled point are quickly found by using kD-tree. Then, the local implicit surface of sampled point that approximates its nearest neighbors is reconstructed based on RBF. Finally, the curvatures of sampled point are estimated by applying the classical differential geometry theory to each implicit surface and their application is given in the point clouds simplification. Some experimental results demonstrated that the method could accurately estimate the curvatures on point sampled model and be effectively applied.%点模型的曲率等微分属性的估算是点模型数字几何处理的基础工作，在点模型数字几何处理的后续工作，如点云简化、特征提取等方面发挥着非常重要的作用。为了较准确地估算点模型的曲率，文章提出一种基于径向基函数（RBF）的点模型曲率估算方法。首先利用kD树，对点模型采样点的最近邻域进行快速搜索；然后基于RBF，对采样点的最近邻域进行局部隐式曲面重构；最后根据经典微分几何理论，在RBF重构曲面上进行曲率估算，同时文章给出了该方法在点云简化中的应用。实验结果表明，该方法对点模型采样点曲率的估算比较精确，并且成功在点云简化中得以应用。
Numerical simulations of 1D inverse heat conduction problems using overdetermined RBF-MLPG method
Ahmad Shirzadi
2013-07-01
Full Text Available This paper proposes a numerical method to deal with the one-dimensional inverse heat conduction problem (IHCP. The initial temperature, a condition on an accessible part of the boundary and an additional temperature measurements in time at an arbitrary location in the domain are known, and it is required to determine the temperature and the heat flux on the remaining part of the boundary. Due to the missing boundary condition, the solution of this problem does not depend continuously on the data and therefore its numerical solution requires special care especially when noise is present in the measured data. In the proposed method, the time variable is eliminated by using finite differences approximation. The method uses a weak formulation of the problem to enjoy the stability condition. To avoid the numerical integration on the whole domain, the weak form equations are constructed on local subdomains. The approximate solution is assumed to be a linear combination of Multi Quadric (MQ radial basis function (RBF constructed on nodal points in the domain and on the boundary. Since the problem is known to be ill-posed, Thikhonov regularization strategy is employed to solve effectively the discrete ill-posed resultant linear system.
RBF-Based Monocular Vision Navigation for Small Vehicles in Narrow Space below Maize Canopy
Lu Liu
2016-06-01
Full Text Available Maize is one of the major food crops in China. Traditionally, field operations are done by manual labor, where the farmers are threatened by the harsh environment and pesticides. On the other hand, it is difficult for large machinery to maneuver in the field due to limited space, particularly in the middle and late growth stage of maize. Unmanned, compact agricultural machines, therefore, are ideal for such field work. This paper describes a method of monocular visual recognition to navigate small vehicles between narrow crop rows. Edge detection and noise elimination were used for image segmentation to extract the stalks in the image. The stalk coordinates define passable boundaries, and a simplified radial basis function (RBF-based algorithm was adapted for path planning to improve the fault tolerance of stalk coordinate extraction. The average image processing time, including network latency, is 220 ms. The average time consumption for path planning is 30 ms. The fast processing ensures a top speed of 2 m/s for our prototype vehicle. When operating at the normal speed (0.7 m/s, the rate of collision with stalks is under 6.4%. Additional simulations and field tests further proved the feasibility and fault tolerance of our method.
李鑫; 杨开明; 朱煜
2012-01-01
According to the modeling error in the dynamical model of robotic manipulators, a new self-adaptive control strategy based on modeling error compensated by RBF neural network was proposed. By means of Computed Torque control method based on inverse dynamics, and through input torque and desired trajectory corrected, two self-adaptive control schemes based on error compensated were developed. The correction terms were learned on-line by RBF neural network, and the adaptive learning law of network weights was developed based on Lyapunov stability theory, therefore the convergence and stability were guaranteed. Simulations are presented for a planner manipulator with two joints, the trajectory tracking results show that the modeling error can be effectively approximated and compensated, the RBF neural network improves the performance and makes the controller have robustness to the parameter perturbations, and it can be applied to the control for robotic manipulators.%针对机械手动力学建模误差,提出了基于RBF神经网络误差补偿的自适应控制策略。在基于逆动力学的计算力矩控制方法的基础上,对系统输入与目标轨迹进行修正,设计了两种误差补偿自适应控制器。利用RBF神经网络对修正项在线自学习,并根据Lyapunov稳定性理论建立了网络权重自适应学习律,保证了跟踪误差的收敛及系统的稳定。以平面转动双臂机械手轨迹跟踪为例进行仿真,结果表明该方法能够有效地补偿建模误差,提高了系统的控制性能并使控制系统具有对参数摄动的鲁棒性,对于机械手自适应控制具有一定的可行性。
王学全; 刘君梅; 杨恒华; 赵学彬; 陈琦
2012-01-01
Evaporation is an important factor affecting thermal balance and water budget over the earth surface. A long-term observation of soil evaporation over semi-fixed dune was carried out with micro-lysimeters （MLS） in the high-frigid regions in the Qinghai-Tibetan Plateau of China during the period of 2006 -2009, the dataset was consisted of the collected daily soil evaporation as the output and the corresponding meteorological observation data including relative air humidity, air temperature, wind speed and soil moisture content as the input. A desert soil evaporation model was developed to research soil evaporation over semi-fixed dune based on the radial basis function （RBF） neural network, and the multiple linear regression （MLR） was used to validate the model. The results show that the values calculated with RBF network output were consistent with the observed values, and the root mean squared error was 0.14 mm. Both the average absolute percent error and the root mean squared error for the RBF neural network were lower than those for the MLR model. The RBF neural network model is good for calculating desert soil evaporation other than the traditional mathematical evaporation model, and it is characterized by the simple development, high accuracy and strong adaptability.%蒸发是地表能量平衡和水分平衡的重要组成部分。2006-2009年在青海东南部沙区半固定沙丘，利用微型蒸发器（MLS）对土壤蒸发进行了测定，结合气象观测数据，利用RBF神经网络技术，建立了沙区半固定沙丘土壤蒸发模型，并应用多元回归技术进行了验证。结果表明：已经构建的RBF神经网络计算土壤蒸发与实测值吻合较好，均方差是0．4mm，其绝对误差和均方差均小于多元线性回归计算值。模型在确定沙区土壤蒸发中具有实用可靠的优势。
Nord, Stefan; Bhatt, Monika J; Tükenmez, Hasan; Farabaugh, Philip J; Wikström, P Mikael
2015-08-01
The in vivo assembly of ribosomal subunits requires assistance by maturation proteins that are not part of mature ribosomes. One such protein, RbfA, associates with the 30S ribosomal subunits. Loss of RbfA causes cold sensitivity and defects of the 30S subunit biogenesis and its overexpression partially suppresses the dominant cold sensitivity caused by a C23U mutation in the central pseudoknot of 16S rRNA, a structure essential for ribosome function. We have isolated suppressor mutations that restore partially the growth of an RbfA-lacking strain. Most of the strongest suppressor mutations alter one out of three distinct positions in the carboxy-terminal domain of ribosomal protein S5 (S5) in direct contact with helix 1 and helix 2 of the central pseudoknot. Their effect is to increase the translational capacity of the RbfA-lacking strain as evidenced by an increase in polysomes in the suppressed strains. Overexpression of RimP, a protein factor that along with RbfA regulates formation of the ribosome's central pseudoknot, was lethal to the RbfA-lacking strain but not to a wild-type strain and this lethality was suppressed by the alterations in S5. The S5 mutants alter translational fidelity but these changes do not explain consistently their effect on the RbfA-lacking strain. Our genetic results support a role for the region of S5 modified in the suppressors in the formation of the central pseudoknot in 16S rRNA.
朱群雄; 李澄非
2006-01-01
Many applications of principal component analysis (PCA) can be found in dimensionality reduction.But linear PCA method is not well suitable for nonlinear chemical processes. A new PCA method based on improved input training neural network (IT-NN) is proposed for the nonlinear system modelling in this paper. Momentum factor and adaptive learning rate are introduced into learning algorithm to improve the training speed of IT-NN. Contrasting to the auto-associative neural network (ANN), IT-NN has less hidden layers and higher training speed. The effectiveness is illustrated through a comparison of IT-NN with linear PCA and ANN with experiments.Moreover, the IT-NN is combined with RBF neural network (RBF-NN) to model the yields of ethylene and propylene in the naphtha pyrolysis system. From the illustrative example and practical application, IT-NN combined with RBF-NN is an effective method of nonlinear chemical process modelling.
赵文秀; 张晓丽; 李国会
2015-01-01
基于我国南方某河流1965—1999年每年7月的实测流量资料，首先采用随机森林模型筛选预报因子，之后利用筛选的预报因子作为RBF神经网络的输入层，利用RBF神经网络对2000—2008年每年7月的流量进行了“滚动式”预报，并与实测结果进行了对比。结果表明：随机森林模型能有效地筛选影响因子，利用这些因子采用RBF神经网络进行径流预报的相对误差均在10％以内，拟合效果很好；“滚动式”长期径流预报结果相对误差的绝对值均在20％以内。%Based on the measured flow data of each July during the period of 1965 to 1999 of a river in southern China,the random forest model was used to filter the impact factors. Then,the RBF network input layer was trained with the selected factors and was utilized to fore-cast the flow data annually in July during 2000—2008 using rolling type pattern. The results show that the random forest model can effective-ly filter out the main factors. Based on the factors filtered by the random forest model,the relative error of RBF networks prediction results is within 10%. The fitting effect is good. Besides that,the relative error of the long-term prediction results using rolling type pattern is within 20%.
RBF network designing based on artificial immune%基于人工免疫系统的RBF网络设计
朱亚男
2015-01-01
由于传统的RBF网络学习方法存在诸多的不足，本文提出基于免疫机制的三级RBF网络学习方法：在第一级得到网络隐层节点数作为疫苗，不仅可自行构建网络，还降低了第二级搜索空间的复杂度；第二级利用人工免疫算法对解空间进行多点搜索，得到全局最优的隐层非线性参数；第三级采用最小二乘法确定网络输出层线性参数，极大地降低了第二级结构的维数，提高了算法效率。经典型Hermit多项式逼近实验验证了该方法训练得到的RBF网络性能优越。%In order to improve the traditional RBF learning strategy, a three-level RBF network learning algorithm based on immune system is proposed, which can calculates the number of the hidden-layer neurons in the first level as immune vaccine, the network can be established and adjusted by itself, and the complexity of search space in the second level can be reduced. The global optimum hidden-layer nonlinear parameters are searched for in the second level by parallel searching with artificial immune algorithm. The output-layer linear parameters are estimated in the third level with least square method, which makes the design dimension of the second level decreased and the algorithm efficiency improved. The experiment of Hermit polynomial approximation shows that the performance of the RBF network trained by the algorithm is superior.
Entropy Generation Due to Natural Convection in a Partially Heated Cavity by Local RBF-DQ Method
Soleimani, S.; Qajarjazi, A.; Bararnia, H.
2011-01-01
The Local Radial Basis Function-Differential Quadrature (RBF-DQ) method is applied to twodimensional incompressible Navier-Stokes equations in primitive form. Numerical results of heatlines and entropy generation due to heat transfer and fluid friction have been obtained for laminar natural...... convection. The variations of the entropy generation for different Rayleigh numbers are also investigated. Comparison between the present results and previous works demonstrated excellent agreements which verify the accuracy and flexibility of the method in simulating the fluid mechanics and heat transfer...
A Radial Basis Function (RBF) Method for the Fully Nonlinear 1D Serre Green-Naghdi Equations
Fabien, Maurice S
2014-01-01
In this paper, we present a spectral method based on Radial Basis Functions (RBFs) for numerically solving the fully nonlinear 1D Serre Green-Naghdi equations. The approximation uses an RBF discretization in space and finite differences in time; the full discretization is obtained by the method of lines technique. For select test cases (see Bonnenton et al. [2] and Kim [11]) the approximation achieves spectral (exponential) accuracy. Complete \\textsc{matlab} code of the numerical implementation is included in this paper (the logic is easy to follow, and the code is under 100 lines).
Barnett, Gregory A; Wicker, Louis J
2015-01-01
Polyharmonic spline (PHS) radial basis functions (RBFs) are used together with polynomials to create local RBF-finite-difference (RBF-FD) weights on different node layouts for spatial discretization of the compressible Navier-Stokes equations at low Mach number, relevant to atmospheric flows. Test cases are taken from the numerical weather prediction community and solved on bounded domains. Thus, attention is given on how to handle boundaries with the RBF-FD method, as well as a novel implementation for the presented approach. Comparisons are done on Cartesian, hexagonal, and quasi-uniformly scattered node layouts. Since RBFs are independent of a coordinate system (and only depend on the distance between nodes), changing the node layout amounts to changing one line of code. In addition, consideration and guidelines are given on PHS order, polynomial degree and stencil size. The main advantages of the present method are: 1) capturing the basic physics of the problem surprisingly well, even at very coarse resol...
LP(K)中RBF神经网络的系统识别问题%LP(K) Approximation Problems in System Identification with RBF Neural Networks
南东; 隆金玲
2009-01-01
Lp approximation problems in system identification with RBF neural networks are investigated.It is proved that by superpositions of some functions of one variable in Lploc(R),one can approximate continuous functionals defined on a compact subset of Lp(K) and continuous operators from a compact subset of Lp1 (K1) to a compact subset of Lp2 (K(2).These results show that if its activation function is in Lploc(R) and is not an even polynomial,then this RBF neural networks can approximate the above systems with any accuracy.
The ATLAS collaboration
2017-01-01
The application of boosted decision trees and deep neural networks to the identification of hadronically-decaying W bosons and top quarks using high-level jet observables as inputs is investigated using Monte Carlo simulations. In the case of both boosted decision trees and deep neural networks, the use of machine learning techniques is found to improve the background rejection with respect to simple reference single jet substructure and mass taggers. Linear correlations between the resulting classifiers and the substructure variables are also presented.
Nitta, Tatsumi; The ATLAS collaboration
2017-01-01
The application of boosted decision trees and deep neural networks to the identification of hadronically-decaying W bosons and top quarks using high-level jet observables as inputs is investigated using Monte Carlo simulations. In the case of both boosted decision trees and deep neural networks, the use of machine learning techniques is found to improve the background rejection with respect to simple reference single jet substructure and mass taggers. Linear correlations between the resulting classifiers and the substructure variables are also presented.
2013-01-01
This is a very simple program to help you put together input files for use in Gries' (2007) R-based collostruction analysis program. It basically puts together a text file with a frequency list of lexemes in the construction and inserts a column where you can add the corpus frequencies. It requires...... it as input for basic collexeme collostructional analysis (Stefanowitsch & Gries 2003) in Gries' (2007) program. ColloInputGenerator is, in its current state, based on programming commands introduced in Gries (2009). Projected updates: Generation of complete work-ready frequency lists....
Combined LTP and LTD of modulatory inputs controls neuronal processing of primary sensory inputs.
Doiron, Brent; Zhao, Yanjun; Tzounopoulos, Thanos
2011-07-20
A hallmark of brain organization is the integration of primary and modulatory pathways by principal neurons. However, the pathway interactions that shape primary input processing remain unknown. We investigated this problem in mouse dorsal cochlear nucleus (DCN) where principal cells integrate primary, auditory nerve input with modulatory, parallel fiber input. Using a combined experimental and computational approach, we show that combined LTP and LTD of parallel fiber inputs to DCN principal cells and interneurons, respectively, broaden the time window within which synaptic inputs summate. Enhanced summation depolarizes the resting membrane potential and thus lowers the response threshold to auditory nerve inputs. Combined LTP and LTD, by preserving the variance of membrane potential fluctuations and the membrane time constant, fixes response gain and spike latency as threshold is lowered. Our data reveal a novel mechanism mediating adaptive and concomitant homeostatic regulation of distinct features of neuronal processing of sensory inputs.
基于RBF神经网络的语音情感识别%Speech Emotion Recognition Based on RBF Neural Network
张海燕; 唐建芳
2011-01-01
The principle of radial base function neural network and its train algorithm are introduced in this paper.Meanwhile,the model of speech emotion recognition based on RBF neural network is established.In the recognition experiments,BP neural network and RBF neural network are compared in the same testing environment.The recognition rate of RBF neural network is 3% more than BP neural network.The results show that the method based on RBF neural network speech emotion recognition is effective.%介绍了径向基函数神经网络的原理、训练算法,并建立了RBF神经网络的语音情感识别的模型。在实验中比较了BP神经网络与RBF神经网络分别用于语音情感识别识别率,RBF神经网络的平均识别率高于BP神经网络3%。结果表明,基于RBF神经网络的语音情感识别方法的有效性。
Inter-provincial mobility features of natural gas input in China%我国省际天然气资源流调入规模分布的分形分析
丛殿阁; 林健宸; 陈孝劲; 赵奎涛; 叶张煌
2015-01-01
This paper studys the size distribution of nature gas input among provinces from 2002 to 2012 based on Zipf law and difference degree model .The conclusion is :① Size distribution of nature gas input among provinces from 2002 to 2012 basically follow Zipf law ,and the change of Zipf parameter is consistent with the change of difference degree .② Provinces in the non-scaling ranges is generally increasing ,which shows size distribution of nature gas input among provinces in China is being gradually optimized .③Fractal evolution law of size distribution of input of nature gas among provinces is different each year .Except 2011 ,each year has two non-scaling ranges ,complying with double fractal .2011 has a non-scaling range ,complying with single fractal .④ In the first range of 2002~2005 ,the rank-size distribution of the nature gas input in this four years complies with lognormal distribution model;2006 ~ 2008 ,Pareto model;2009~2012 ,lognormal distribution model .In the second range of each year except 2011 ,the rank-size distribution in those years complies with irregular Pareto model .⑤ The change of size distribution of the nature gas input was relative to many factors ,e .g .the production place ,reserves and yield of large gas field , discovery of new gas field and decline of old gas field ,level of economic development among different provinces and years .%本文运用了位次变化、齐夫定律、差异度 ,比较分析了2000~2012年我国省际天然气资源流调入规模分布的变化特征.得出以下结论 :从各省历年天然气调入量位次变化看出近年来我国天然气资源流调入规模分布趋于分散 ,呈良性趋势发展 ;研究时段内我国省际天然气资源流调入规模结构都基本遵循齐夫法则 ,而且与差异度变化趋势也是一致的.这说明齐夫法则可为我国省际天然气资源流调入规模结构变化的研究提供新的定量方法.从省际天然气调入量来看 ,无标
Automatic solar feature detection using image processing and pattern recognition techniques
Qu, Ming
The objective of the research in this dissertation is to develop a software system to automatically detect and characterize solar flares, filaments and Corona Mass Ejections (CMEs), the core of so-called solar activity. These tools will assist us to predict space weather caused by violent solar activity. Image processing and pattern recognition techniques are applied to this system. For automatic flare detection, the advanced pattern recognition techniques such as Multi-Layer Perceptron (MLP), Radial Basis Function (RBF), and Support Vector Machine (SVM) are used. By tracking the entire process of flares, the motion properties of two-ribbon flares are derived automatically. In the applications of the solar filament detection, the Stabilized Inverse Diffusion Equation (SIDE) is used to enhance and sharpen filaments; a new method for automatic threshold selection is proposed to extract filaments from background; an SVM classifier with nine input features is used to differentiate between sunspots and filaments. Once a filament is identified, morphological thinning, pruning, and adaptive edge linking methods are applied to determine filament properties. Furthermore, a filament matching method is proposed to detect filament disappearance. The automatic detection and characterization of flares and filaments have been successfully applied on Halpha full-disk images that are continuously obtained at Big Bear Solar Observatory (BBSO). For automatically detecting and classifying CMEs, the image enhancement, segmentation, and pattern recognition techniques are applied to Large Angle Spectrometric Coronagraph (LASCO) C2 and C3 images. The processed LASCO and BBSO images are saved to file archive, and the physical properties of detected solar features such as intensity and speed are recorded in our database. Researchers are able to access the solar feature database and analyze the solar data efficiently and effectively. The detection and characterization system greatly improves
Waste treatment in physical input-output analysis
Dietzenbacher, E
2005-01-01
When compared to monetary input-output tables (MIOTs), a distinctive feature of physical input-output tables (PIOTs) is that they include the generation of waste as part of a consistent accounting framework. As a consequence, however, physical input-output analysis thus requires that the treatment o
Zhang, Ridong; Tao, Jili; Lu, Renquan; Jin, Qibing
2016-12-08
Modeling of distributed parameter systems is difficult because of their nonlinearity and infinite-dimensional characteristics. Based on principal component analysis (PCA), a hybrid modeling strategy that consists of a decoupled linear autoregressive exogenous (ARX) model and a nonlinear radial basis function (RBF) neural network model are proposed. The spatial-temporal output is first divided into a few dominant spatial basis functions and finite-dimensional temporal series by PCA. Then, a decoupled ARX model is designed to model the linear dynamics of the dominant modes of the time series. The nonlinear residual part is subsequently parameterized by RBFs, where genetic algorithm is utilized to optimize their hidden layer structure and the parameters. Finally, the nonlinear spatial-temporal dynamic system is obtained after the time/space reconstruction. Simulation results of a catalytic rod and a heat conduction equation demonstrate the effectiveness of the proposed strategy compared to several other methods.
WANG Hong-qi; WANG Xue-yuan; TANG Yu
2005-01-01
This paper designs an intelligent evaluation approach using a Radial Basis Function (RBF) Artificial Neural Network. We based our approach on establishing a comprehensive advantage evaluating index system that offers scientific substance for creating a development plan and the strategic management of high-tech industry and regional cluslers of high-tech enterprises. Furthermore, this paper selects some typical high-tech enterprises' data to make comprehensive training on the network system. Meanwhile, the paper chooses some enterprises as testing samples to test the method, the result of which proves that this method is truly effective. The research of this paper provides a comprehensive advantage evaluating and managing method for high-tech enterprise.
郭小燕; 张明
2013-01-01
针对RBF神经网络确定核函数中心时没有考虑输入样本分类指标权重的问题,提出了一种动态加权聚类算法.在算法中利用样本之间的加权距离代替了欧氏距离作为选定核函数中心的量度.在此基础上,建立了信用评价模型,利用已知类别的样本对模型进行训练,再利用训练好的模型对未知类别的样本进行预测,实验结果验证了模型的有效性.%A dynamic weighting cluster algorithm is proposed in this article in view of the problem of input sample's classification weight being not considered by formerly RBF neural network. In this algorithm, the weighting distance replaces the Euclidean distance to act the role of measurement to the cluster. Based on this, the credit evaluation model is established, which is trained by known category sample. Then the trained model is used to forecast the unknown category sample, the experimental result confirms the model' s validity.
Soft sensor for ratio of soda to aluminate based on PCA-RBF multiple network
GUI Wei-hua; LI Yong-gang; WANG Ya-lin
2005-01-01
Based on principal component analysis, a multiple neural network was proposed. The principal component analysis was firstly used to reorganize the input variables and eliminate the correlativity. Then the reorganized variables were divided into 2 groups according to the original information and 2 corresponding neural networks were established. A radial basis function network was used to depict the relationship between the output variables and the first group input variables which contain main original information. An other single-layer neural network model was used to compensate the error between the output of radial basis function network and the actual output variables. At last, The multiple network was used as soft sensor for the ratio of soda to aluminate in the process of high-pressure digestion of alumina. Simulation of industry application data shows that the prediction error of the model is less than 3%, and the model has good generalization ability.
李楠; 赵均海; 王娟; 吴赛
2014-01-01
针对混杂纤维混凝土强度受多种因素影响，强度与各影响因素之间关系为复杂的非线性问题，通过人工神经网络的自适应、自学习和非线性映射，可以找到以影响因素为输入变量、以混杂纤维混凝土强度为输出变量之间的非线性关系，在文献试验实测值的基础上采用MATLAB神经网络工具箱建立了四个三层RBF和BP神经网络模型，采用所建立的RBF和BP神经网络对混杂纤维混凝土的抗拉强度和抗折强度分别进行预测，并将各自的预测值和实测值进行了对比分析。结果表明：RBF神经网络预测值与试验实测值吻合良好，较之BP神经网络有更高的强度预测能力，该方法可行且预测精度满足工程需要，为工程上研究混杂纤维混凝土强度提供了新方法。%The strength of hybrid fiber reinforced concrete is influenced by many factors,and the relationship between them are complex nonlinear problem,but the nonlinear relationship between input variables like some of the factors and output variables like the strength of hybrid fiber reinforced concrete can be obtained by self-adapting,self-studying and nonlinear mapping of artificial neural network.Based on experimental values,four RBF and BP neural network models were established in MATLAB neural network toolbox,compressive strength and flexural strength of hybrid fiber reinforced concrete were predicted respectively by using RBF and BP neural network model. The predicted values and measured values were analyzed in comparison.The results showed that the predicted values of RBF neural network was in good agreement with the experimental values,and compared with the BP neural network had a higher strength prediction ability,the method was feasible and the prediction accuracy can meet the needs of engineering,providing a new method for the research on strength of hybrid fiber reinforced concrete in engineering field.
Scaling of global input-output networks
Liang, Sai; Qi, Zhengling; Qu, Shen; Zhu, Ji; Chiu, Anthony S. F.; Jia, Xiaoping; Xu, Ming
2016-06-01
Examining scaling patterns of networks can help understand how structural features relate to the behavior of the networks. Input-output networks consist of industries as nodes and inter-industrial exchanges of products as links. Previous studies consider limited measures for node strengths and link weights, and also ignore the impact of dataset choice. We consider a comprehensive set of indicators in this study that are important in economic analysis, and also examine the impact of dataset choice, by studying input-output networks in individual countries and the entire world. Results show that Burr, Log-Logistic, Log-normal, and Weibull distributions can better describe scaling patterns of global input-output networks. We also find that dataset choice has limited impacts on the observed scaling patterns. Our findings can help examine the quality of economic statistics, estimate missing data in economic statistics, and identify key nodes and links in input-output networks to support economic policymaking.
蒋加伏; 赵怡
2015-01-01
The traditional human action recognition algorithm tends to focus on solving a certain behavior recognition,it cannot be generalized.So,this paper put forward a kind of Local evidence RBF algorithm based high-level characteristic self similarity fusion for human behavior recognition.Firstly,the time-dependent generalized self similarity concept and the spa-tio-temporal interest point optical flow based local features extraction method were used to construct the human behavior description based on self similar matrix.Secondly,after independent individual behavior recognition in the use of SVM algo-rithm,the evidence theory based high level feature fusion was used to realize the optimization for classification of structure, which can improve the accuracy of classification.Simulation results show that the proposed scheme can significantly improve the efficiency and accuracy for human action recognition.%针对传统人体动作识别算法，往往重点解决某一类行为识别，不具有通用性的问题，提出一种局部证据 RBF人体行为高层特征自相似融合识别算法。首先，借用随时间变化的广义自相似性概念，利用时空兴趣点光流场局部特征提取方法，构建基于自相似矩阵的人体行为局部特征描述；其次，在使用 SVM算法进行独立个体行为识别后，利用所提出的证据理论 RBF(Radial Basis Function)高层特征融合，实现分类结构优化，从而提高分类准确度；仿真实验表明，所提方案能够明显提高人体行为识别算法效率和识别准确率。
Selecting training inputs via greedy rank covering
Buchsbaum, A.L.; Santen, J.P.H. van [AT& T Bell Laboratories, Murray Hill, NJ (United States)
1996-12-31
We present a general method for selecting a small set of training inputs, the observations of which will suffice to estimate the parameters of a given linear model. We exemplify the algorithm in terms of predicting segmental duration of phonetic-segment feature vectors in a text-to-speech synthesizer, but the algorithm will work for any linear model and its associated domain.
Feature selection in bioinformatics
Wang, Lipo
2012-06-01
In bioinformatics, there are often a large number of input features. For example, there are millions of single nucleotide polymorphisms (SNPs) that are genetic variations which determine the dierence between any two unrelated individuals. In microarrays, thousands of genes can be proled in each test. It is important to nd out which input features (e.g., SNPs or genes) are useful in classication of a certain group of people or diagnosis of a given disease. In this paper, we investigate some powerful feature selection techniques and apply them to problems in bioinformatics. We are able to identify a very small number of input features sucient for tasks at hand and we demonstrate this with some real-world data.
Nondestructive diagnosis of flip chips based on vibration analysis using PCA-RBF
Su, Lei; Shi, Tielin; Liu, Zhiping; Zhou, Hongdi; Du, Li; Liao, Guanglan
2017-02-01
Flip chip technology combined with solder bump interconnection has been widely applied in IC package. The solder bumps are sandwiched between dies and substrates, leading to conventional techniques being difficult to diagnose the flip chips. Meanwhile, these conventional diagnosis methods are usually performed by human visual judgment. The human eye-fatigue can easily cause fault detection. Thus, it is difficult and crucial to detect the defects of flip chips automatically. In this paper, a nondestructive diagnosis system based on vibration analysis is proposed. The flip chip is excited by air-coupled ultrasounds and raw vibration signals are measured by a laser scanning vibrometer. Forty-two features are extracted for analysis, including ten time domain features, sixteen frequency domain features and sixteen wavelet packet energy features. Principal component analysis is used for feature reduction. Radial basis function neural network is adopted for classification and recognition. Flip chips in three states (good flip chips, flip chips with missing solder bumps and flip chips with open solder bumps) are utilized to validate the proposed method. The results demonstrate that this method is effective for defect inspection in flip chip package.
李慧; 徐文尚; 李迪; 刘杰; 孙运营
2012-01-01
The RBF neural network was chosen to build pH value and concentration models and to adjust reac tor solution so that slow convergence speed and relative minimum and other weakness of negative gradient de scent method can be controlled. Actual application shows that both speed and precision can meet the require ment of pH value and concentration control.%选用径向基函数神经网络建立pH值和浓度的模型,克服了在白炭黑生产过程中采用负梯度下降法调节反应釜溶液pH值和浓度时存在收敛速度慢和局部极小等缺点.实际应用表明:其速度和精度完全达到了工艺上对pH值和浓度的控制要求.
Shankar, Varun; Wright, Grady B.; Fogelson, Aaron L.; Kirby, Robert M.
2014-05-01
We present a computational method for solving the coupled problem of chemical transport in a fluid (blood) with binding/unbinding of the chemical to/from cellular (platelet) surfaces in contact with the fluid, and with transport of the chemical on the cellular surfaces. The overall framework is the Augmented Forcing Point Method (AFM) (\\emph{L. Yao and A.L. Fogelson, Simulations of chemical transport and reaction in a suspension of cells I: An augmented forcing point method for the stationary case, IJNMF (2012) 69, 1736-52.}) for solving fluid-phase transport in a region outside of a collection of cells suspended in the fluid. We introduce a novel Radial Basis Function-Finite Difference (RBF-FD) method to solve reaction-diffusion equations on the surface of each of a collection of 2D stationary platelets suspended in blood. Parametric RBFs are used to represent the geometry of the platelets and give accurate geometric information needed for the RBF-FD method. Symmetric Hermite-RBF interpolants are used for enforcing the boundary conditions on the fluid-phase chemical concentration, and their use removes a significant limitation of the original AFM. The efficacy of the new methods are shown through a series of numerical experiments; in particular, second order convergence for the coupled problem is demonstrated.
RBF网络在有源降噪系统中的应用仿真%The Application Simulation of RBF Neural Network in Active Noise Control System
姜丽飞
2011-01-01
针对车辆舱室内的噪声特点,在分析噪声系统非线性特性的基础上,将高速实时信号处理器DSP应用于有源降噪系统的硬件设计中,提出一种基于变结构RBF神经网络的噪声自适应控制方案,给出滤波-x算法并进行仿真验证.同时将该滤波算法的降噪效果和基本的滤波-x算法的进行比较.结果表明,采用该控制算法取得了良好的降噪效果.%Aiming at the characteristics of noise in vehicle cabins, and analyzing nonlinear characteristics of the system, the high-speed real-time signal processor (DSP) was used in active noise control system and an adaptive noise control project based on RBF networks was proposed.An algorithm of FX-RBF was given and simulated.Its noise reduction effect was compared with that of the basic filter-X algorithm.Simulation results showed that the noise reduction effect of the FX-RBF algorithm is good.
Goudarzi, Nasser
2016-04-05
In this work, two new and powerful chemometrics methods are applied for the modeling and prediction of the (19)F chemical shift values of some fluorinated organic compounds. The radial basis function-partial least square (RBF-PLS) and random forest (RF) are employed to construct the models to predict the (19)F chemical shifts. In this study, we didn't used from any variable selection method and RF method can be used as variable selection and modeling technique. Effects of the important parameters affecting the ability of the RF prediction power such as the number of trees (nt) and the number of randomly selected variables to split each node (m) were investigated. The root-mean-square errors of prediction (RMSEP) for the training set and the prediction set for the RBF-PLS and RF models were 44.70, 23.86, 29.77, and 23.69, respectively. Also, the correlation coefficients of the prediction set for the RBF-PLS and RF models were 0.8684 and 0.9313, respectively. The results obtained reveal that the RF model can be used as a powerful chemometrics tool for the quantitative structure-property relationship (QSPR) studies.
Rough RBF Neural Network Based on Extreme Learning%基于极速学习的粗糙RBF神经网络
马刚; 丁世飞; 史忠植
2012-01-01
提出了一种用于训练粗糙RBF神经网络（rough RBF neural networks,R-RBF）的极速学习机（extreme learning machine,ELM）方法,通过引入矩阵的Moore-Penrose逆,将传统的迭代学习方法转换为一种求线性方程的极小范数最小二乘解的方法.实验证明,在训练精度、训练时间上都能够达到非常优越的性能,其泛化精度能够提升50%以上.%The paper proposes a method of training rough RBF neural networks（R-RBF） using the extreme learning machine（ELM）, which eonverts the traditional iterative training method to solve norm least-squares solution of general linear system by introducing Moore-Penrose inverse. Experiments show that it can reach a very superior performance in both time and aeeuraey when ELM trains the Rough RBF Neural Networks, which can improve the generalization accuracy more than 50% compared with the traditional thinking of adjusting parameters iterative[y.
Judit Navracsics
2014-01-01
Full Text Available According to the critical period hypothesis, the earlier the acquisition of a second language starts, the better. Owing to the plasticity of the brain, up until a certain age a second language can be acquired successfully according to this view. Early second language learners are commonly said to have an advantage over later ones especially in phonetic/phonological acquisition. Native-like pronunciation is said to be most likely to be achieved by young learners. However, there is evidence of accentfree speech in second languages learnt after puberty as well. Occasionally, on the other hand, a nonnative accent may appear even in early second (or third language acquisition. Cross-linguistic influences are natural in multilingual development, and we would expect the dominant language to have an impact on the weaker one(s. The dominant language is usually the one that provides the largest amount of input for the child. But is it always the amount that counts? Perhaps sometimes other factors, such as emotions, ome into play? In this paper, data obtained from an EnglishPersian-Hungarian trilingual pair of siblings (under age 4 and 3 respectively is analyzed, with a special focus on cross-linguistic influences at the phonetic/phonological levels. It will be shown that beyond the amount of input there are more important factors that trigger interference in multilingual development.
A linear-RBF multikernel SVM to classify big text corpora.
Romero, R; Iglesias, E L; Borrajo, L
2015-01-01
Support vector machine (SVM) is a powerful technique for classification. However, SVM is not suitable for classification of large datasets or text corpora, because the training complexity of SVMs is highly dependent on the input size. Recent developments in the literature on the SVM and other kernel methods emphasize the need to consider multiple kernels or parameterizations of kernels because they provide greater flexibility. This paper shows a multikernel SVM to manage highly dimensional data, providing an automatic parameterization with low computational cost and improving results against SVMs parameterized under a brute-force search. The model consists in spreading the dataset into cohesive term slices (clusters) to construct a defined structure (multikernel). The new approach is tested on different text corpora. Experimental results show that the new classifier has good accuracy compared with the classic SVM, while the training is significantly faster than several other SVM classifiers.
王云静; 李燕; 曲正伟; 刘圣楠
2016-01-01
It is of great significance to accurately identify the type of disturbance signal to analyze and control the pow-er quality problem.In this paper, a new method of power quality disturbance identification based on matching pursuit optimized by particle swarm optimization ( PSO-MP) and RBF neural network is proposed.Firstly, in order to let the residue signal to better reflect the different disturbance signal difference, the fundamental atomic library is constructed to extract the fundamental frequency signals;Then, the MP algorithm is optimized by PSO to reduce the calculation a-mount, which combines with discrete Gabor atom libraries to accurately extract atomic parameters of residual disturb-ance signal by sparse decomposition, Finally, the RBF neural network is used to identify disturbance signals by fea-tures, which is the mean and standard deviation of the atomic parameter and projection of residual signal on the atom. Simulation examples show that the proposed method can effectively identify several common power quality disturbances with a small amount of computation and good anti-noise performance.%准确识别扰动信号类型对分析和治理电能质量问题具有重要意义。文中提出一种基于粒子群优化匹配追踪算法（ PSO－MP）和RBF神经网络的电能质量扰动识别方法。首先，构建工频原子库将工频信号提取出来，得到的残余信号能更好地体现扰动信号差异性；再利用PSO优化匹配追踪算法以减小计算量，并结合离散Gabor原子库对残余扰动信号进行稀疏分解，准确提取其原子参数；最后将原子参数以及残余信号在原子上的投影的均值和标准偏差作为特征量，利用RBF神经网络对扰动信号进行识别。仿真算例表明，该方法能够有效地识别几种常见的电能质量扰动，且具有抗噪性能强、计算量小等优点。
A Deep Web Query Interfaces Classification Method Based on RBF Neural Network
YUAN Fang; ZHAO Yao; ZHOU Xu
2007-01-01
This paper proposes a new approach for classification for query interfaces of Deep Web, which extracts features from the form's text data on the query interfaces, assisted with the synonym library, and uses radial basic function neural network (RBFNN) algorithm to classify the query interfaces. The applied RBFNN is a kind of effective feed-forward artificial neural network, which has a simple networking structure but features with strength of excellent nonlinear approximation, fast convergence and global convergence. A TEL_8 query interfaces' data set from UIUC on-line database is used in our experiments, which consists of 477 query interfaces in 8 typical domains. Experimental results proved that the proposed approach can efficiently classify the query interfaces with an accuracy of 95.67%.
Goudarzi, Shidrokh; Haslina Hassan, Wan; Abdalla Hashim, Aisha-Hassan; Soleymani, Seyed Ahmad; Anisi, Mohammad Hossein; Zakaria, Omar M.
2016-01-01
This study aims to design a vertical handover prediction method to minimize unnecessary handovers for a mobile node (MN) during the vertical handover process. This relies on a novel method for the prediction of a received signal strength indicator (RSSI) referred to as IRBF-FFA, which is designed by utilizing the imperialist competition algorithm (ICA) to train the radial basis function (RBF), and by hybridizing with the firefly algorithm (FFA) to predict the optimal solution. The prediction accuracy of the proposed IRBF–FFA model was validated by comparing it to support vector machines (SVMs) and multilayer perceptron (MLP) models. In order to assess the model’s performance, we measured the coefficient of determination (R2), correlation coefficient (r), root mean square error (RMSE) and mean absolute percentage error (MAPE). The achieved results indicate that the IRBF–FFA model provides more precise predictions compared to different ANNs, namely, support vector machines (SVMs) and multilayer perceptron (MLP). The performance of the proposed model is analyzed through simulated and real-time RSSI measurements. The results also suggest that the IRBF–FFA model can be applied as an efficient technique for the accurate prediction of vertical handover. PMID:27438600
PENG Lei; LIU Li; LONG Teng; GUO Xiaosong
2014-01-01
As a promising technique, surrogate-based design and optimization(SBDO) has been widely used in modern engineering design optimizations. Currently, static surrogate-based optimization methods have been successfully applied to expensive optimization problems. However, due to the low efficiency and poor flexibility, static surrogate-based optimization methods are difficult to efficiently solve practical engineering cases. At the aim of enhancing efficiency, a novel surrogate-based efficient optimization method is developed by using sequential radial basis function(SEO-SRBF). Moreover, augmented Lagrangian multiplier method is adopted to solve the problems involving expensive constraints. In order to study the performance of SEO-SRBF, several numerical benchmark functions and engineering problems are solved by SEO-SRBF and other well-known surrogate-based optimization methods including EGO, MPS, and IARSM. The optimal solutions, number of function evaluations, and algorithm execution time are recorded for comparison. The comparison results demonstrate that SEO-SRBF shows satisfactory performance in both optimization efficiency and global convergence capability. The CPU time required for running SEO-SRBF is dramatically less than that of other algorithms. In the torque arm optimization case using FEA simulation, SEO-SRBF further reduces 21% of thematerial volume compared with the solution from static-RBF subject to the stress constraint. This study provides the efficient strategy to solve expensive constrained optimization problems.
权值与结构双确定法的RBF神经网络分类器%RBF Neural Network Classifier with Weights and Structure Determination Method
张雨浓; 王茹; 廖柏林; 刘锦荣; 林键煜
2014-01-01
In order to solve the difficulties in determining the weights and structure of the radial basis function (RBF) neural network.Based on the weights-direct-determination (WDD)method and the relationship among centers,variances, the number of hidden-layer neurons and the performance of the neural network,a pruning-while-growing-type weights-and-structure-determination (PWGT-WASD)algorithm is proposed.On the basis of the PWGT-WASD algorithm,a kind of RBF neural network classifier is constructed,and its classifying and antinoise ability is further discussed in this paper.Com-puter numerical experiment results substantiate that the proposed PWGT-WASD algorithm can determine the centers,the va-riances and the optimal weights and structure of RBF neural network quickly and effectively.The constructed RBF pattern classifier has the superiority in terms of classification and antinoise ability.%为了解决径向基函数(RBF)神经网络权值与结构难以确定的问题，基于权值直接确定法，及隐层神经元中心、方差、数目与神经网络性能的关系，提出一种边增边删型的网络权值与结构双确定法。在此方法基础之上，构建一种 RBF神经网络分类器并探讨其分类性能和抗噪能力。计算机数值实验结果验证所提出的边增边删型的权值与结构双确定法能够快速、有效地确定网络的中心、方差和网络最优的权值与结构，所构造的模式分类器具有优越的分类性能和抗噪能力。
Czarnitzki, Dirk; Grimpe, Christoph; Pellens, Maikel
The viability of modern open science norms and practices depend on public disclosure of new knowledge, methods, and materials. However, increasing industry funding of research can restrict the dissemination of results and materials. We show, through a survey sample of 837 German scientists in life...... sciences, natural sciences, engineering, and social sciences, that scientists who receive industry funding are twice as likely to deny requests for research inputs as those who do not. Receiving external funding in general does not affect denying others access. Scientists who receive external funding...... of any kind are, however, 50% more likely to be denied access to research materials by others, but this is not affected by being funded specifically by industry....
Czarnitzki, Dirk; Grimpe, Christoph; Pellens, Maikel
2015-01-01
The viability of modern open science norms and practices depends on public disclosure of new knowledge, methods, and materials. However, increasing industry funding of research can restrict the dissemination of results and materials. We show, through a survey sample of 837 German scientists in life...... sciences, natural sciences, engineering, and social sciences, that scientists who receive industry funding are twice as likely to deny requests for research inputs as those who do not. Receiving external funding in general does not affect denying others access. Scientists who receive external funding...... of any kind are, however, 50 % more likely to be denied access to research materials by others, but this is not affected by being funded specifically by industry...
Eye movement identification based on accumulated time feature
Guo, Baobao; Wu, Qiang; Sun, Jiande; Yan, Hua
2017-06-01
Eye movement is a new kind of feature for biometrical recognition, it has many advantages compared with other features such as fingerprint, face, and iris. It is not only a sort of static characteristics, but also a combination of brain activity and muscle behavior, which makes it effective to prevent spoofing attack. In addition, eye movements can be incorporated with faces, iris and other features recorded from the face region into multimode systems. In this paper, we do an exploring study on eye movement identification based on the eye movement datasets provided by Komogortsev et al. in 2011 with different classification methods. The time of saccade and fixation are extracted from the eye movement data as the eye movement features. Furthermore, the performance analysis was conducted on different classification methods such as the BP, RBF, ELMAN and SVM in order to provide a reference to the future research in this field.
Fire Risk Assessment of Some Indian Coals Using Radial Basis Function (RBF) Technique
Nimaje, Devidas; Tripathy, Debi Prasad
2016-03-01
Fires, whether surface or underground, pose serious and environmental problems in the global coal mining industry. It is causing huge loss of coal due to burning and loss of lives, sterilization of coal reserves and environmental pollution. Most of the instances of coal mine fires happening worldwide are mainly due to the spontaneous combustion. Hence, attention must be paid to take appropriate measures to prevent occurrence and spread of fire. In this paper, to evaluate the different properties of coals for fire risk assessment, forty-nine in situ coal samples were collected from major coalfields of India. Intrinsic properties viz. proximate and ultimate analysis; and susceptibility indices like crossing point temperature, flammability temperature, Olpinski index and wet oxidation potential method of Indian coals were carried out to ascertain the liability of coal to spontaneous combustion. Statistical regression analysis showed that the parameters of ultimate analysis provide significant correlation with all investigated susceptibility indices as compared to the parameters of proximate analysis. Best correlated parameters (ultimate analysis) were used as inputs to the radial basis function network model. The model revealed that Olpinski index can be used as a reliable method to assess the liability of Indian coals to spontaneous combustion.
Fire Risk Assessment of Some Indian Coals Using Radial Basis Function (RBF) Technique
Nimaje, Devidas; Tripathy, Debi Prasad
2017-04-01
Fires, whether surface or underground, pose serious and environmental problems in the global coal mining industry. It is causing huge loss of coal due to burning and loss of lives, sterilization of coal reserves and environmental pollution. Most of the instances of coal mine fires happening worldwide are mainly due to the spontaneous combustion. Hence, attention must be paid to take appropriate measures to prevent occurrence and spread of fire. In this paper, to evaluate the different properties of coals for fire risk assessment, forty-nine in situ coal samples were collected from major coalfields of India. Intrinsic properties viz. proximate and ultimate analysis; and susceptibility indices like crossing point temperature, flammability temperature, Olpinski index and wet oxidation potential method of Indian coals were carried out to ascertain the liability of coal to spontaneous combustion. Statistical regression analysis showed that the parameters of ultimate analysis provide significant correlation with all investigated susceptibility indices as compared to the parameters of proximate analysis. Best correlated parameters (ultimate analysis) were used as inputs to the radial basis function network model. The model revealed that Olpinski index can be used as a reliable method to assess the liability of Indian coals to spontaneous combustion.
Automated Stellar Classification for Large Surveys with EKF and RBF Neural Networks
Ling Bai; Ping Guo; Zhan-Yi Hu
2005-01-01
An automated classification technique for large size stellar surveys is proposed. It uses the extended Kalman filter as a feature selector and pre-classifier of the data, and the radial basis function neural networks for the classification.Experiments with real data have shown that the correct classification rate can reach as high as 93%, which is quite satisfactory. When different system models are selected for the extended Kalman filter, the classification results are relatively stable. It is shown that for this particular case the result using extended Kalman filter is better than using principal component analysis.
Mao-long Lv
2016-01-01
Full Text Available In the process of ultralow altitude airdrop, many factors such as actuator input dead-zone, backlash, uncertain external atmospheric disturbance, and model unknown nonlinearity affect the precision of trajectory tracking. In response, a robust adaptive neural network dynamic surface controller is developed. As a result, the aircraft longitudinal dynamics with actuator input nonlinearity is derived; the unknown nonlinear model functions are approximated by means of the RBF neural network. Also, an adaption strategy is used to achieve robustness against model uncertainties. Finally, it has been proved that all the signals in the closed-loop system are bounded and the tracking error converges to a small residual set asymptotically. Simulation results demonstrate the perfect tracking performance and strong robustness of the proposed method, which is not only applicable to the actuator with input dead-zone but also suitable for the backlash nonlinearity. At the same time, it can effectively overcome the effects of dead-zone and the atmospheric disturbance on the system and ensure the fast track of the desired flight path angle instruction, which overthrows the assumption that system functions must be known.
Latham, D. W.; Brown, T. M.; Monet, D. G.; Everett, M.; Esquerdo, G. A.; Hergenrother, C. W.
2005-12-01
The Kepler mission will monitor 170,000 planet-search targets during the first year, and 100,000 after that. The Kepler Input Catalog (KIC) will be used to select optimum targets for the search for habitable earth-like transiting planets. The KIC will include all known catalogued stars in an area of about 177 square degrees centered at RA 19:22:40 and Dec +44:30 (l=76.3 and b=+13.5). 2MASS photometry will be supplemented with new ground-based photometry obtained in the SDSS g, r, i, and z bands plus a custom filter centered on the Mg b lines, using KeplerCam on the 48-inch telescope at the Whipple Observatory on Mount Hopkins, Arizona. The photometry will be used to estimate stellar characteristics for all stars brighter than K 14.5 mag. The KIC will include effective temperature, surface gravity, metallicity, reddening, distance, and radius estimates for these stars. The CCD images are pipeline processed to produce instrumental magnitudes at PSI. The photometry is then archived and transformed to the SDSS system at HAO, where the astrophysical analysis of the stellar characteristics is carried out. The results are then merged with catalogued data at the USNOFS to produce the KIC. High dispersion spectroscopy with Hectochelle on the MMT will be used to supplement the information for many of the most interesting targets. The KIC will be released before launch for use by the astronomical community and will be available for queries over the internet. Support from the Kepler mission is gratefully acknowledged.
ABC optimized RBF network for classification of EEG signal for epileptic seizure identification
Sandeep Kumar Satapathy
2017-03-01
Full Text Available The brain signals usually generate certain electrical signals that can be recorded and analyzed for detection in several brain disorder diseases. These small signals are expressly called as Electroencephalogram (EEG signals. This research work analyzes the epileptic disorder in human brain through EEG signal analysis by integrating the best attributes of Artificial Bee Colony (ABC and radial basis function networks (RBFNNs. We have used Discrete Wavelet Transform (DWT technique for extraction of potential features from the signal. In our study, for classification of these signals, in this paper, the RBFNNs have been trained by a modified version of ABC algorithm. In the modified ABC, the onlooker bees are selected based on binary tournament unlike roulette wheel selection of ABC. Additionally, kernels such as Gaussian, Multi-quadric, and Inverse-multi-quadric are used for measuring the effectiveness of the method in numerous mixtures of healthy segments, seizure-free segments, and seizure segments. Our experimental outcomes confirm that RBFNN with inverse-multi-quadric kernel trained with modified ABC is significantly better than RBFNNs with other kernels trained by ABC and modified ABC.
Waite, Anthony; /SLAC
2011-09-07
Serial Input/Output (SIO) is designed to be a long term storage format of a sophistication somewhere between simple ASCII files and the techniques provided by inter alia Objectivity and Root. The former tend to be low density, information lossy (floating point numbers lose precision) and inflexible. The latter require abstract descriptions of the data with all that that implies in terms of extra complexity. The basic building blocks of SIO are streams, records and blocks. Streams provide the connections between the program and files. The user can define an arbitrary list of streams as required. A given stream must be opened for either reading or writing. SIO does not support read/write streams. If a stream is closed during the execution of a program, it can be reopened in either read or write mode to the same or a different file. Records represent a coherent grouping of data. Records consist of a collection of blocks (see next paragraph). The user can define a variety of records (headers, events, error logs, etc.) and request that any of them be written to any stream. When SIO reads a file, it first decodes the record name and if that record has been defined and unpacking has been requested for it, SIO proceeds to unpack the blocks. Blocks are user provided objects which do the real work of reading/writing the data. The user is responsible for writing the code for these blocks and for identifying these blocks to SIO at run time. To write a collection of blocks, the user must first connect them to a record. The record can then be written to a stream as described above. Note that the same block can be connected to many different records. When SIO reads a record, it scans through the blocks written and calls the corresponding block object (if it has been defined) to decode it. Undefined blocks are skipped. Each of these categories (streams, records and blocks) have some characteristics in common. Every stream, record and block has a name with the condition that each
基于PSO-RBF的建筑能耗预测模型研究%Prediction Model of Building Energy Consumption Based on PSO-RBF
季文娟; 顾永松
2015-01-01
Themodelofenergyconsumptionpredictionisbuiltafteranalyzingcharacteristicson energy consumption changes of public building in hot summer and cold winter area. Particle swarm optimization algorithm is used to optimize the model, and the PSO-RBF neural network prediction model is established. Using the energy consumption data of subject research, the samples of building energy consumption is built. Then the RBF neural network and PSO-RBF neural network are trained on MATLAB. Experiments are conducted to predict energy consumption values of typical public buildings. The results show that accuracy of the prediction model is improved obviously after being optimized, and it has strong learning and predicting ability. The model can predict energy consumption value of public buildings accurately.%通过研究分析夏热冬冷地区公共建筑能耗变化特点, 建立了RBF神经网络建筑能耗预测模型. 在此基础上运用微粒群算法对模型优化,建立了基于PSO-RBF的建筑能耗预测模型. 利用大量数据构造样本集,运用软件分别对优化前后的预测模型进行训练,并运用到典型公共建筑能耗值的预测实例中. 结果表明基于PSO-RBF的建筑能耗预测模型的学习能力和预测能力强,能较准确地实现公共建筑能耗预测.
利用RBF神经网络实现高斯型函数积分%Implementation for Gauss- Type Function Integral Using RBF Neural Networks
杨军; 马晓岩; 万山虎; 江晶
2003-01-01
导出了在一定精度下高斯型函数积分近似表达式,利用径向基函数(RBF)网络具有良好的逼近任意非线性映射的特点,提出了一种改进的RBF网络方法以实现对高斯型函数积分.实验结果表明所提出方法具有较高的逼近计算精度.
Application of RBF neural network to fault diagnosis in heliostats filed%RBF神经网络在定日镜场故障诊断中的应用
王成昱; 万定生; 郭铁铮
2011-01-01
针对定日镜场故障与征兆之间的关系特点,介绍了RBF神经网络运用于定日镜场故障诊断的基本方法.利用MATLAB神经网络工具箱建立和训练RBF神经网络,并对网络进行了测试.结果说明RBF神经网络在定目镜场故障诊断中具有较高的准确性和可靠性.%For the characteristic of the relationship between faults and symptoms, the basic principle and method of application of RBF neural network technique for the fault diagnosis in heliostats filed were introduced. The RBF neural network was built by using the neural network toolbox of MATLAB. The test result showed the use of the RBF network neural model was accurate and reliable.
Inhibitory Gating of Input Comparison in the CA1 Microcircuit.
Milstein, Aaron D; Bloss, Erik B; Apostolides, Pierre F; Vaidya, Sachin P; Dilly, Geoffrey A; Zemelman, Boris V; Magee, Jeffrey C
2015-09-23
Spatial and temporal features of synaptic inputs engage integration mechanisms on multiple scales, including presynaptic release sites, postsynaptic dendrites, and networks of inhibitory interneurons. Here we investigate how these mechanisms cooperate to filter synaptic input in hippocampal area CA1. Dendritic recordings from CA1 pyramidal neurons reveal that proximal inputs from CA3 as well as distal inputs from entorhinal cortex layer III (ECIII) sum sublinearly or linearly at low firing rates due to feedforward inhibition, but sum supralinearly at high firing rates due to synaptic facilitation, producing a high-pass filter. However, during ECIII and CA3 input comparison, supralinear dendritic integration is dynamically balanced by feedforward and feedback inhibition, resulting in suppression of dendritic complex spiking. We find that a particular subpopulation of CA1 interneurons expressing neuropeptide Y (NPY) contributes prominently to this dynamic filter by integrating both ECIII and CA3 input pathways and potently inhibiting CA1 pyramidal neuron dendrites.
Input in Second Language Acquisition.
Gass, Susan M., Ed.; Madden, Carolyn G., Ed.
This collection of conference papers includes: "When Does Teacher Talk Work as Input?"; "Cultural Input in Second Language Learning"; "Skilled Variation in a Kindergarten Teacher's Use of Foreigner Talk"; "Teacher-Pupil Interaction in Second Language Development"; "Foreigner Talk in the University…
Saleemi, Anjum P.
1989-01-01
Major approaches of describing or examining linguistic data from a potential target language (input) are analyzed for adequacy in addressing the concerns of second language learning theory. Suggestions are made for making the best of these varied concepts of input and for reformulation of a unified concept. (MSE)
Input and Second Language Acquisition
周笑盈
2011-01-01
The behaviorist, the mentalist and the interactionist have different emphases on the role input in Second Language Acquisition. In order to protrude the importance of second language teaching, it is indispensible to discuss the characteristics of input and to explore its effects.
Approach to red tide prediction on RBF neural network%基于RBF神经网络的赤潮预测方法
李慧; 顾沈明
2012-01-01
Red tide is an anomalous phenomenon and is characterized by abruptness and nonlinearity, so the red tide prediction has been a hotspot in the oceanographic studies. The fundamentals of RBF neural network are briefly introduced, and the application of artificial neural network method to the red tide prediction is discussed. Based on the RBF neural network, the simulation experiments are also illustrated by using red tide monitoring data, and the experimental analysis is also proposed.%赤潮是一种由多因素综合作用引发的生态异常现象,具有突发性及非线性等特点.对其进行预测预报一直是海洋科学研究的热点.简要介绍了RBF神经网络的基本原理,探讨了应用该人工神经网络进行赤潮预测的方法.利用RBF神经网络模型对赤潮灾害监测数据进行仿真实验,并对结果进行了分析.
ATTACK-DEFENSE GAME MODEL BASED ON RBF NEURAL NETWORK%基于RBF神经网络的攻防博弈模型
娄燕强; 宋如顺; 马永彩
2011-01-01
In order to elucidate how to determine the types between game sides in the process of network attack and defense and then to choose the action strategy, the attack - defense game model based on RBF neural network is put forward. Firstly, the two - player stochastic game model is used to analyze the characteristics of offensive and defensive sides, to reveal the restrict factors of strategies selection. Then,the optimal strategies chosen by both sides can be got through perfect Bayesian Nash equilibrium. At last, the RBF neural network is adopted to reason out the types of the suspects according to their action strategies and system status.%为了阐明网络攻防过程中博弈双方如何确定对方的类型,从而选择行动策略,提出了基于RBF神经网络的攻防博弈模型.首先使用两人随机博弈模型来分析网络攻防双方的特点,揭示制约双方选择策略的因素;通过精炼贝叶斯纳什均衡求得博弈双方选择的最优策略;最后,根据可疑者的行动策略和系统的状况,使用RBF神经网络对其类型进行推理.
Research on Telephone Charge Estimate Based on RBF Neural Network%基于RBF神经网络的话费估计问题研究
孙立炜; 林峰
2013-01-01
Due to the expense of the lost data,it is important to estimate the lost data according to the existing data. In this paper,the lost telephone charging data are estimated by RBF neural network,which achieves better effect. When RBF neural network is applied to numerical value estimating,it is important to conclude principles of choosing the best estimate value form aspects of the spread constant,nerve cells number and mean squared error.%数据丢失常常会造成损失，需要根据已有数据估计所丢失的数据。本文利用RBF神经网络估计丢失的话费数据，取得了较好的效果。在利用RBF神经网络处理数值估计问题时，要从散布常数、神经元个数和均方误差三个方面归纳最优估计值选择原则。
Input management of production systems.
Odum, E P
1989-01-13
Nonpoint sources of pollution, which are largely responsible for stressing regional and global life-supporting atmosphere, soil, and water, can only be reduced (and ultimately controlled) by input management that involves increasing the efficiency of production systems and reducing the inputs of environmentally damaging materials. Input management requires a major change, an about-face, in the approach to management of agriculture, power plants, and industries because the focus is on waste reduction and recycling rather than on waste disposal. For large-scale ecosystem-level situations a top-down hierarchical approach is suggested and illustrated by recent research in agroecology and landscape ecology.
Input-output-controlled nonlinear equation solvers
Padovan, Joseph
1988-01-01
To upgrade the efficiency and stability of the successive substitution (SS) and Newton-Raphson (NR) schemes, the concept of input-output-controlled solvers (IOCS) is introduced. By employing the formal properties of the constrained version of the SS and NR schemes, the IOCS algorithm can handle indefiniteness of the system Jacobian, can maintain iterate monotonicity, and provide for separate control of load incrementation and iterate excursions, as well as having other features. To illustrate the algorithmic properties, the results for several benchmark examples are presented. These define the associated numerical efficiency and stability of the IOCS.
Enhanced feature integration in musicians
Hansen, Niels Christian; Højlund, Andreas; Møller, Cecilie
Distinguishing and integrating features of sensory input is essential to human survival and no less paramount in music perception and cognition. Yet, little is known about training-induced plasticity of neural mechanisms for auditory feature integration. This study aimed to contrast the two...
The Advanced LIGO Input Optics
Mueller, Chris; Ciani, Giacomo; DeRosa, Ryan; Effler, Anamaria; Feldbaum, David; Frolov, Valery; Fulda, Paul; Gleason, Joseph; Heintze, Matthew; King, Eleanor; Kokeyama, Keiko; Korth, William; Martin, Rodica; Mullavey, Adam; Poeld, Jan; Quetschke, Volker; Reitze, David; Tanner, David; Williams, Luke; Mueller, Guido
2016-01-01
The Advanced LIGO gravitational wave detectors are nearing their design sensitivity and should begin taking meaningful astrophysical data in the fall of 2015. These resonant optical interferometers will have unprecedented sensitivity to the strains caused by passing gravitational waves. The input optics play a significant part in allowing these devices to reach such sensitivities. Residing between the pre-stabilized laser and the main interferometer, the input optics is tasked with preparing the laser beam for interferometry at the sub-attometer level while operating at continuous wave input power levels ranging from 100 mW to 150 W. These extreme operating conditions required every major component to be custom designed. These designs draw heavily on the experience and understanding gained during the operation of Initial LIGO and Enhanced LIGO. In this article we report on how the components of the input optics were designed to meet their stringent requirements and present measurements showing how well they h...
An evaluation of cardiorespiratory and movement features with respect to sleep-stage classification.
Willemen, T; Van Deun, D; Verhaert, V; Vandekerckhove, M; Exadaktylos, V; Verbraecken, J; Van Huffel, S; Haex, B; Sloten, J Vander
2014-03-01
Polysomnography (PSG) is considered the gold standard to assess sleep accurately, but it can be expensive, time-consuming, and uncomfortable, specifically in long-term sleep studies. Actigraphy, on the other hand, is both cheap and userfriendly, but depending on the application lacks detail and accuracy. Our aim was to evaluate cardiorespiratory and movement signals in discriminating between wake, rapid-eye-movement (REM), light (N1N2), and deep (N3) sleep. The dataset comprised 85 nights of PSG from a healthy population. Starting from a total of 750 characteristic variables (features), problem-specific subsets of 40 features were forwardly selected using the combination of a wrapper method (Cohen's kappa statistic on radial basis function (RBF)-kernel support vector machine (SVM) classifier) and filter method (minimum redundancy maximum relevance criterion on mutual information). Final classification was performed using an RBF-kernel SVM. Non-subject-specific wake versus sleep classification resulted in a Cohen’s kappa value of 0.695, while REM versus NREM resulted in 0.558 and N3 versus N1N2 in 0.553. The broad pool of initial features gave insight in which features discriminated best between the different classes. The classification results demonstrate the possibility of making long-term sleep monitoring more widely available.
Nonlinear input-output systems
Hunt, L. R.; Luksic, Mladen; Su, Renjeng
1987-01-01
Necessary and sufficient conditions that the nonlinear system dot-x = f(x) + ug(x) and y = h(x) be locally feedback equivalent to the controllable linear system dot-xi = A xi + bv and y = C xi having linear output are found. Only the single input and single output case is considered, however, the results generalize to multi-input and multi-output systems.
J. W. Horng
2012-04-01
Full Text Available Two current conveyors (CCs based high input impedance voltage-mode universal biquadratic filters each with three input terminals and one output terminal are presented. The first circuit is composed of three differential voltage current conveyors (DVCCs, two grounded capacitors and four resistors. The second circuit is composed of two DVCCs, one differential difference current conveyor (DDCC, two grounded capacitors and four grounded resistors. The proposed circuits can realize all the standard filter functions, namely, lowpass, bandpass, highpass, notch and allpass filters by the selections of different input voltage terminals. The proposed circuits offer the features of high input impedance, using only grounded capacitors and low active and passive sensitivities. Moreover, the x ports of the DVCCs (or DDCC in the proposed circuits are connected directly to resistors. This design offers the feature of a direct incorporation of the parasitic resistance at the x terminal of the DVCC (DDCC, Rx, as a part of the main resistance.
A Practical pedestrian approach to parsimonious regression with inaccurate inputs
Seppo Karrila
2014-04-01
Full Text Available A measurement result often dictates an interval containing the correct value. Interval data is also created by roundoff, truncation, and binning. We focus on such common interval uncertainty in data. Inaccuracy in model inputs is typically ignored on model fitting. We provide a practical approach for regression with inaccurate data: the mathematics is easy, and the linear programming formulations simple to use even in a spreadsheet. This self-contained elementary presentation introduces interval linear systems and requires only basic knowledge of algebra. Feature selection is automatic; but can be controlled to find only a few most relevant inputs; and joint feature selection is enabled for multiple modeled outputs. With more features than cases, a novel connection to compressed sensing emerges: robustness against interval errors-in-variables implies model parsimony, and the input inaccuracies determine the regularization term. A small numerical example highlights counterintuitive results and a dramatic difference to total least squares.
Multiple input/output random vibration control system
Unruh, James F.
1988-01-01
A multi-input/output random vibration control algorithm was developed based on system identification concepts derived from random vibration spectral analysis theory. The unique features of the algorithm are: (1) the number of input excitors and the number of output control responses need not be identical; (2) the system inverse response matrix is obtained directly from the input/output spectral matrix; and (3) the system inverse response matrix is updated every control loop cycle to accommodate system amplitude nonlinearities. A laboratory demonstration case of two imputs with three outputs is presented to demonstrate the system capabilities.
[Prosody, speech input and language acquisition].
Jungheim, M; Miller, S; Kühn, D; Ptok, M
2014-04-01
In order to acquire language, children require speech input. The prosody of the speech input plays an important role. In most cultures adults modify their code when communicating with children. Compared to normal speech this code differs especially with regard to prosody. For this review a selective literature search in PubMed and Scopus was performed. Prosodic characteristics are a key feature of spoken language. By analysing prosodic features, children gain knowledge about underlying grammatical structures. Child-directed speech (CDS) is modified in a way that meaningful sequences are highlighted acoustically so that important information can be extracted from the continuous speech flow more easily. CDS is said to enhance the representation of linguistic signs. Taking into consideration what has previously been described in the literature regarding the perception of suprasegmentals, CDS seems to be able to support language acquisition due to the correspondence of prosodic and syntactic units. However, no findings have been reported, stating that the linguistically reduced CDS could hinder first language acquisition.
王瑞; 史天运; 王彤
2011-01-01
对实测风速数据进行Kalman滤波,去除实测风速数据的偏差；通过归一化处理,消除数据中的冗余成分;针对RBF神经网络的预测误差会随着时间的推移而增大的问题,采用滚动式训练方法在线训练RBF神经网络;用训练好的RBF神经网络进行风速预测,再对预测结果进行反归一化处理,得到最终的预测风速.仿真结果表明,运用基于RBF神经网络的铁路短时风速预测方法对短时风速进行预测,最大相对误差仅为5.92％,可满足铁路防灾安全监控系统中风速预测子系统的要求.%The measured wind speed was processed with Kalman filter algorithm to eliminate deviations. The redundancies in the measured data were removed through normalization processing. Then, RBF neural network was online trained by using the rolling training method to deal with the problem that the prediction error of RBF neural network would increase as time went on. Finally, the wind speed was predicted by using the well-trained RBF neural network. The final forecasted wind speed was then obtained by anti-normalizing the output of RBF neural network. The simulation results show that the maximum relative error is only 5. 92% using the proposed railway short-time wind speed prediction algorithm based on RBF neural network, which can satisfy the requirements of the wind forecasting subsystem in railway disaster prevention and safety monitoring system.
基于RBF神经网络的自适应均衡器研究%Study on New Adaptive Equalizers Based on RBF Neural Networks
王军锋; 褚晓勇; 宋国乡
2002-01-01
在研究基于径向基函数(RBF)神经网络的均衡器结构以及传统自适应均衡算法的基础上,提出了两种新的基于RBF神经网络的自适应均衡器,并给出了相应的自适应均衡算法.新的均衡器是将判决反馈引入到RBF神经网络中以及将Adaline网络与RBF网络有机的结合而分别构成的,仿真结果表明这两种新算法比基于RBF神经网络的自适应均衡算法都具有更好的收敛性能.
陈红杰; 李高锋
2015-01-01
针对建筑工程特点，提出了基于RBF神经网络的建筑工程投标报价方法，建立建筑工程投标报价标高率数学模型。应用MATLAB计算软件，以实例验证了该模型的正确性及实用性。%According to the peculiarity of construction engineering, based on the theory of RBF neural networks. Mark-up rate evaluation model of construction engineering is set up. A particular software MATLAB is employed to analyze the evaluation model. The accuracy and value of the approach are testified by the prototypical data.
Human Reliability Prediction of Lifting Operation Based on RBF Neural Network%基于RBF的起重作业岗位人因可靠性预测
王洪德; 马成正
2012-01-01
In order to improve the reliability of lifting operation, and prevent accident caused by human errors, bearing randomness, fuzziness and uncertainty of human error in mind, an RBF neural network-based method for analyzing the human error's nonlinear dynamics process was put forwarded. Taking the lifting operation as example, firstly, an indexes system about the human reliability prediction was constructed , which included the factors of the operator, the communion interface, the operating circumstance, the operating characteristics and the operating organization. Then the indexes were quantified. Secondly, according to human reliability analysis (HRA) theory and the scene record, the human error data were calculated out, and the human error rates were given. Finally, basing on the analysis of the operator's tiredness and emotion, information channels, operation complexity and time margin, lighting and wind power conditions, working pressure and safety supervision, an RBF neural network-based model for lifting operation human reliability was built. The results indicate that the RBF prediction includes the operation reliability as well as the cognitive reliability, and that the predictions results conform with the actual observed values up to 92. 0%.%为提高起重作业可靠性,防止人因失误酿成事故,针对人因失误的随机性、模糊性和不确定性特点,提出运用具有非线性映射能力和容错能力的径向基函数( RBF)神经网络,分析人因失误非线性动力学过程.以起重机操作岗位作为人因可靠性分析(HRA)实例,首先,建立基于“作业人员、交流界面、作业环境、作业特性、作业组织”的人因可靠性预测指标体系,并对指标进行量化；其次,根据人因可靠性原理,统计出人因失误次数,给出人因失误率；最后,通过对“人的疲劳和情绪、交流通道、作业复杂程度和时间裕度、照明环境和风力影响、工作强度和安全监管”等
Input Processing and Processing Instruction: Definitions and Issues
Hossein Hashemnezhad
2013-01-01
Full Text Available Input Processing (IP proposed by VanPatten (1993, was innovated based on Krashen’s (1982 input hypothesis. In IP model, principles are stated that describe how learners either miss grammatical markers in the input or how they get them wrong (VanPatten, 2002b. Based on this model, learners process input for meaning before form. Processing Instruction (PI, an explicit focus on form that is informed by the model of IP, is a practical solution to IP model. The goal of PI is to help L2 learners derive richer intake from input by having them engage in structured input activities that push them away from the strategies they normally use to make form-meaning connections (Wong, 2004. This article intends to study the definitions of IP and PI as well the issues of IP and PI, including the principles of IP, features and goal of PI, and input used in PI (Structured input activities, and then to introduce difference between the terms IP and PI.
COGNITIVE INTERPRETATION OF INPUT HYPOTHESIS
WangHongyue; RenLiankui
2004-01-01
Krashen's Input Hypothesis, together with its earlier version, the Monitor Model is an influential theory in Second Language Acquisition research. In his studies, Krashen, on the one hand, emphasizes the part '“ comprehensible input” plays in learning a second language, on the other hand, he simply defines“comprehensible input” as “a little beyond the learner's current level”. What input can be considered as“a little beyond the learner's current level ”? Krashen gives no furtherexplanation. This paper tries to offer a more concrete and more detailed interpretation with Ausubel's Cognitive Assimilation theory.
Input Hypothesis and its Controversy
金灵
2016-01-01
With Krashen's proposal of input hypothesis in 1980s, lots of contributions and further researches have been done in second language acquisition and teaching. Since it is impossible to undertake the exact empirical research to investigate its credibility, lots of criticisms are also aroused to disprove or adjust this hypothesis. However, due to its significant development in SLA, it is still valuable to explore the hypothesis and implications in language teaching to non-native speakers. This paper firstly focuses on the development of the input hypothesis, and then discusses some criticisms of this hypothesis.
一种改进的RBF神经网络参数优化方法%Improved method for RBF neural network parameters optimization
张辉; 柴毅
2012-01-01
An improved method for RBF neural network parameters optimization is proposed. The number of nodes in the hidden layer is determined by using RAN (Resource Allocating Network), meanwhile strategy of pruning is introduced to remove those hidden units which make insignificant contribution to overall network output. Central position, width and weight of the neural network are optimized by the improved PSO (Particle Swarm Optimization) algorithm, so as to obtain the appropriate structure and control parameters. The new algorithm is used to predict the model of CSTR, and the result indicates that RBF neural network optimized by this algorithm has a smaller structure and high generalization ability.%提出了一种改进的RBF神经网络参数优化算法.通过资源分配网络算法确定隐含层节点个数,引入剪枝策略删除对网络贡献不大的节点,用改进的粒子群算法对RBF网络的中心、宽度、权值进行优化,使RBF网络不仅可以得到合适的结构,同时也可以得到合适的控制参数.将此算法用于连续搅拌釜反应器模型的预测,结果表明,此算法优化后的RBF网络结构小,并且具有较高的泛化能力.
ZHANG Hao
2017-08-01
Full Text Available With SiO2 as the carrier, decanoic acid-palmitic acid as a phase change material,the micron SiO2-based phase change and humidity controlling composite materials were prepared by sol-gel method. The scheme was optimized by uniform design in a combination with RBF neural network to optimizing preparation of micron SiO2-based phase change and humidity controlling composite materials. The performance of micron SiO2-based phase change and humidity controlling composite materials with optimal uniform particle size distribution were tested and characterized. The results show that RBF neural network has the best approximation effect, when spread is 0.5; optimization technology parameters are solution pH value 4.27, amount of deionized water (mole ratio between deionized water and tetraethyl orthosilicate is 8.58, amount of absolute alcohol (mole ratio between absolute alcohol and tetraethyl orthosilicate is 4.83 and ultrasonic wave power is 316W; micron SiO2-based phase change and humidity controlling composite materials with optimal uniform particle size distribution' d10 is 383.51nm, d50 is 511.63nm and d90 is 658.76nm, measured value of d90-d10 is 275.25nm, the measured value and the predicted value are in good agreement (relative error is -2.64%; micron SiO2-based phase change and humidity controlling composite materials with optimal uniform particle size distribution' equilibrium moisture content in the relative humidity of 40%-60% is 0.0925-0.1493g/g, phase transition temperature is 20.02-23.45℃ and phase change enthalpy is 54.06-60.78J/g.
Input-Based Approaches to Teaching Grammar: A Review of Classroom-oriented Research
Rod Ellis
2006-01-01
@@ 2. Input-processing studies The input-processing studies examined here involve experimental comparisons of input-based and productionbased instruction. Whereas traditional grammar teaching attempts to manipulate learner production, ‘processing instruction’ emplys ‘interpretation tasks’ (Ellis 1995) that invite learners to engage in intentional learning by consciously noticing how a target grammatical feature is used in input specially contrived to contain numerous exemplars of the structure.
Federica Cerina
Full Text Available Production systems, traditionally analyzed as almost independent national systems, are increasingly connected on a global scale. Only recently becoming available, the World Input-Output Database (WIOD is one of the first efforts to construct the global multi-regional input-output (GMRIO tables. By viewing the world input-output system as an interdependent network where the nodes are the individual industries in different economies and the edges are the monetary goods flows between industries, we analyze respectively the global, regional, and local network properties of the so-called world input-output network (WION and document its evolution over time. At global level, we find that the industries are highly but asymmetrically connected, which implies that micro shocks can lead to macro fluctuations. At regional level, we find that the world production is still operated nationally or at most regionally as the communities detected are either individual economies or geographically well defined regions. Finally, at local level, for each industry we compare the network-based measures with the traditional methods of backward linkages. We find that the network-based measures such as PageRank centrality and community coreness measure can give valuable insights into identifying the key industries.
On Adaptive Optimal Input Design
Stigter, J.D.; Vries, D.; Keesman, K.J.
2003-01-01
The problem of optimal input design (OID) for a fed-batch bioreactor case study is solved recursively. Here an adaptive receding horizon optimal control problem, involving the so-called E-criterion, is solved on-line, using the current estimate of the parameter vector at each sample instant {tk, k =
Cerina, Federica; Zhu, Zhen; Chessa, Alessandro; Riccaboni, Massimo
2015-01-01
Production systems, traditionally analyzed as almost independent national systems, are increasingly connected on a global scale. Only recently becoming available, the World Input-Output Database (WIOD) is one of the first efforts to construct the global multi-regional input-output (GMRIO) tables. By viewing the world input-output system as an interdependent network where the nodes are the individual industries in different economies and the edges are the monetary goods flows between industries, we analyze respectively the global, regional, and local network properties of the so-called world input-output network (WION) and document its evolution over time. At global level, we find that the industries are highly but asymmetrically connected, which implies that micro shocks can lead to macro fluctuations. At regional level, we find that the world production is still operated nationally or at most regionally as the communities detected are either individual economies or geographically well defined regions. Finally, at local level, for each industry we compare the network-based measures with the traditional methods of backward linkages. We find that the network-based measures such as PageRank centrality and community coreness measure can give valuable insights into identifying the key industries. PMID:26222389
Input in an Institutional Setting.
Bardovi-Harlig, Kathleen; Hartford, Beverly S.
1996-01-01
Investigates the nature of input available to learners in the institutional setting of the academic advising session. Results indicate that evidence for the realization of speech acts, positive evidence from peers and status unequals, the effect of stereotypes, and limitations of a learner's pragmatic and grammatical competence are influential…
Optimal Inputs for System Identification.
1995-09-01
The derivation of the power spectral density of the optimal input for system identification is addressed in this research. Optimality is defined in...identification potential of general System Identification algorithms, a new and efficient System Identification algorithm that employs Iterated Weighted Least
Analog Input Data Acquisition Software
Arens, Ellen
2009-01-01
DAQ Master Software allows users to easily set up a system to monitor up to five analog input channels and save the data after acquisition. This program was written in LabVIEW 8.0, and requires the LabVIEW runtime engine 8.0 to run the executable.
1972-01-01
A general view of the remote input/output station installed in building 112 (ISR) and used for submitting jobs to the CDC 6500 and 6600. The card reader on the left and the line printer on the right are operated by programmers on a self-service basis.
Cerina, Federica; Zhu, Zhen; Chessa, Alessandro; Riccaboni, Massimo
2015-01-01
Production systems, traditionally analyzed as almost independent national systems, are increasingly connected on a global scale. Only recently becoming available, the World Input-Output Database (WIOD) is one of the first efforts to construct the global multi-regional input-output (GMRIO) tables. By viewing the world input-output system as an interdependent network where the nodes are the individual industries in different economies and the edges are the monetary goods flows between industries, we analyze respectively the global, regional, and local network properties of the so-called world input-output network (WION) and document its evolution over time. At global level, we find that the industries are highly but asymmetrically connected, which implies that micro shocks can lead to macro fluctuations. At regional level, we find that the world production is still operated nationally or at most regionally as the communities detected are either individual economies or geographically well defined regions. Finally, at local level, for each industry we compare the network-based measures with the traditional methods of backward linkages. We find that the network-based measures such as PageRank centrality and community coreness measure can give valuable insights into identifying the key industries.
Ozyazici, E. M.
1980-01-01
Module detects level changes in any of its 16 inputs, transfers changes to its outputs, and generates interrupts when changes are detected. Up to four changes-in-state per line are stored for later retrieval by controlling computer. Using standard TTL logic, module fits 19-inch rack-mounted console.
The advanced LIGO input optics
Mueller, Chris L., E-mail: cmueller@phys.ufl.edu; Arain, Muzammil A.; Ciani, Giacomo; Feldbaum, David; Fulda, Paul; Gleason, Joseph; Heintze, Matthew; Martin, Rodica M.; Reitze, David H.; Tanner, David B.; Williams, Luke F.; Mueller, Guido [University of Florida, Gainesville, Florida 32611 (United States); DeRosa, Ryan T.; Effler, Anamaria; Kokeyama, Keiko [Louisiana State University, Baton Rouge, Louisiana 70803 (United States); Frolov, Valery V.; Mullavey, Adam [LIGO Livingston Observatory, Livingston, Louisiana 70754 (United States); Kawabe, Keita; Vorvick, Cheryl [LIGO Hanford Observatory, Richland, Washington 99352 (United States); King, Eleanor J. [University of Adelaide, Adelaide, SA 5005 (Australia); and others
2016-01-15
The advanced LIGO gravitational wave detectors are nearing their design sensitivity and should begin taking meaningful astrophysical data in the fall of 2015. These resonant optical interferometers will have unprecedented sensitivity to the strains caused by passing gravitational waves. The input optics play a significant part in allowing these devices to reach such sensitivities. Residing between the pre-stabilized laser and the main interferometer, the input optics subsystem is tasked with preparing the laser beam for interferometry at the sub-attometer level while operating at continuous wave input power levels ranging from 100 mW to 150 W. These extreme operating conditions required every major component to be custom designed. These designs draw heavily on the experience and understanding gained during the operation of Initial LIGO and Enhanced LIGO. In this article, we report on how the components of the input optics were designed to meet their stringent requirements and present measurements showing how well they have lived up to their design.
龙亿; 杜志江; 王伟东
2015-01-01
为改善外骨骼机器人灵敏度放大控制( SAC)性能,结合遗传算法( GA)与径向基函数( RBF)神经网络建立在线计算外骨骼机器人的精确动力学模型.用GA优化RBF神经网络的中心矢量与基宽度,并对RBF网络的权值实时更新,在线学习外骨骼机器人动力学模型中的参数矩阵,进一步推导出SAC控制律.仿真结果表明:GA优化后的RBF网络,可以在线学习外骨骼的动力学模型,基于该模型的SAC能够实现精确的人体轨迹跟踪,相比于优化前,人体轨迹跟踪误差以及人机交互信息会快速减小并收敛到0的微小邻域内,可实现人机协调运动.%To improve performance of sensitivity amplification control ( SAC ) for exoskeleton robot, genetic algorithm( GA) and RBF neural network was combined to obtain accurate dynamic model of exoskeleton robot online. Parameters of center vector and base width of RBF neural network were obtained by GA optimization, and online RBF weights learning process was constructed to obtain estimation matrix parameters of dynamics system, finally, SAC control law was deduced. Simulation results showed that the RBF network optimized by GA could learn exoskeleton dynamics model parameters online. Based on the learned model, the SAC could achieve more precise human trajectory tracking where tracking error and human-robot interaction force converged to the small neighborhood of zero simultaneously compared with those without optimization. The proposed RBF neural network with GA optimization which learned dynamics model parameters online for exoskeleton robot dynamics model could achieve highly accurate trajectory following for SAC, ultimately realize cooperative movement between human and exoskeleton.
Classification of Broken Rice Kernels using 12D Features
SUNDER ALI KHOWAJA
2016-07-01
Full Text Available Integrating the technological aspect for assessment of rice quality is very much needed for the Asian markets where rice is one of the major exports. Methods based on image analysis has been proposed for automated quality assessment by taking into account some of the textural features. These features are good at classifying when rice grains are scanned in controlled environment but it is not suitable for practical implementation. Rice grains are placed randomly on the scanner which neither maintains the uniformity in intensity regions nor the placement strategy is kept ideal thus resulting in false classification of grains. The aim of this research is to propose a method for extracting set of features which can overcome the said issues. This paper uses morphological features along-with gray level and Hough transform based features to overcome the false classification in the existing methods. RBF (Radial Basis function is used as a classification mechanism to classify between complete grains and broken grains. Furthermore the broken grains are classified into two classes? i.e. acceptable grains and non-acceptable grains. This research also uses image enhancement technique prior to the feature extraction and classification process based on top-hat transformation. The proposed method has been simulated in MATLAB to visually analyze and validate the results.
Optocoupled line receiver input discriminates against narrow noise pulses
Napier, T M
1977-01-01
Describes a simple optocoupled interface which connects a data line to the receiving end of a data link that features pulse length discrimination to enhance noise pulse rejection. A rugged red LED, D /sub 1/, can bypass any reasonable fault currents to protect the relatively fragile optocoupler input diode. (0 refs).
Study of Chunks Input Approach
马静
2003-01-01
This paper is to describe and investigate Chunks (Lexical Phrases ) Input Approach in vocabulary learning strategies by means of achievement tests,questionnaire surveys and interviews. The study is to reveal how different learners combine different vocabulary learning strategies in their learning process. With the data collected, the author of this paper discusses and summarizes learners' individual differences in selecting vocabulary learning strategies with the hope of giving new insights into English teaching and learning.
黄榕波; 郭穗勋
2010-01-01
提出变量可分离函数的径向基函数网络拟合模型(Fitting Model based Radial Basis Function network to Variable Separable Function,VSRBF)及其学习算法并分析VSRBF的VC维.VSRBF是一个由多个子径向基函数网络组成的分工协作系统,由于把高维模型分解为低维模型,与传统径向基函数网络(Based Radial Basis Function Network,RBF)相比, VSRBF 不仅明显地降低了系统复杂性而且网络的收敛速度更快. 证明了VSRBF的VC维低于传统RBF的VC维,实验表明VSRBF在处理高维模型的行为明显优于RBF.
Structure of RBF with long and short term memory based on field knowledge%基于先验知识的长短记忆RBF网络结构
韩丽; 史丽萍; 徐治皋
2008-01-01
提出了一种基于先验知识的RBF-LSFM(RBF with Long and Short Term Memory)网络结构.该网络将专业背景知识引入到神经网络的结构构造中,提出了具有长短期记忆功能的网络结构.同时引入了剪枝理论,使网络具有更精简的结构.将这种网络结构应用于热工过程中过热气温动态特性建模,仿真结果表明该神经网络模型具有较高的建模精度以及泛化能力.
Understanding Legacy Features with Featureous
Olszak, Andrzej; Jørgensen, Bo Nørregaard
2011-01-01
Feature-centric comprehension of source code is essential during software evolution. However, such comprehension is oftentimes difficult to achieve due the discrepancies between structural and functional units of object-oriented programs. We present a tool for feature-centric analysis of legacy...
吕岚; 甘旭升; 屈虹; 赵海涛
2014-01-01
为提高RBF神经网络的建模性能，提出一种基于改进无迹Kalman滤波（UKF）的RBF神经网络训练算法。在该算法中，首先将比例最小偏度单形Sigma点采样策略引入UT，以有效改进UKF，提升其计算效率，然后利用改进的UKF优化估计RBF神经网络的最优参数。仿真结果表明，改进的UKF比EKF具有更高的RBF神经网络模型训练精度，与传统UKF的模型精度大体相当，但速度更快，计算效率更高。%To improve the modeling of RBF neural network,a training algorithm of RBF neural network based on modified Unscented Kalman Filter(UKF)is proposed. In the algorithm,first a scaled minimal skew simplex Sigma point sampling strategy is introduced in Unscented Transform (UT) to improve UKF for computation efficiency,and then the improved UKF is used to optimize the parameters of RBF neural network. Simulation show that,for the training problem of RBF neural network,the model precision of proposed UKF is higher than that of EKF,and is approximately close to traditional UKF with faster training and better computation.
袁红春; 潘金晶
2016-01-01
In order to improve accuracy of dissolved oxygen prediction,radial basis function( RBF)neural network based on improved recursive least square algorithm is applied to predict the dissolved oxygen. Using K means clustering algorithm to choose the center of hidden layer units and improved recursive least square algorithm is used to optimize the weights of hidden layer to output layer of RBF neural network. Simulation results show that the proposed method has good nonlinear fitting ability and its prediction precision is higher than RBF neural network and RBF neural network based on recusive least square algorithm.%为提高溶解氧预测的准确性，将基于改进型递归最小二乘算法优化的径向基函数（ RBF）神经网络方法应用于溶解氧预测。利用K均值聚类算法进行隐层单元中心选择；利用改进型递归最小二乘算法优化RBF神经网络隐含层到输出层的权值。仿真结果表明：该方法对溶解氧的预测具有较好的非线性拟合能力，预测精度优于RBF神经网络和递归最小二乘算法优化的RBF神经网络。
改进的双模型结构RBF神经网络及其应用%Improved RBF neural network with double model structure and its application
李全善; 张义山; 曹柳林; 林晓琳; 崔佳
2011-01-01
提出了离线结构学习和在线权值校正相结合的双模型结构RBF神经网络,以离线学习和在线校正相结合的方式实现网络的自学习和自校正,满足了软测量仪表现场应用的要求.针对应用过程中出现预测误差过大的现象,通过对网络算法进行分析,研究影响网络预测精度的因素,在此基础上,提出了以K均值聚类法和递推下降算法相结合的RBF神经网络建模改进算法,仿真结果和实际应用证明了改进算法的有效性.%A dual model RBF (radial basis function) neural network was proposed in this paper. One is used for self-learning, which learns one time a day. The other is used for on-line correcting, which is the running model currently. Both the self-learning model and the on-line correcting model are corrected six times every day and should track the current conditions of the system quickly. At the same time, the accuracy of the two models should be compared. If the accuracy of the on-line correcting model is less than the one of the self-learning model, the latter becomes the new currently running model instead of the old one. Otherwise, the currently model is maintained. To solve the problem of neural network large prediction errors, a network algorithm analysis is given and the influence factors of the network prediction accuracy are found. At last, an improved algorithm of RBF neural network modeling is proposed, which combines K-means clustering method with the recursive descent algorithm. Simulation and practical application proved the effectiveness of the improved method.
U.S. Environmental Protection Agency — This dataset consists of various site features from multiple Superfund sites in U.S. EPA Region 8. These data were acquired from multiple sources at different times...
CERN. Geneva
2015-01-01
Feature selection and reduction are key to robust multivariate analyses. In this talk I will focus on pros and cons of various variable selection methods and focus on those that are most relevant in the context of HEP.
National Oceanic and Atmospheric Administration, Department of Commerce — Collection includes a variety of solar feature datasets contributed by a number of national and private solar observatories located worldwide.
连黎明; 唐军
2014-01-01
通过对数控转台SKZT3500用恒流静压轴承进行特性分析，借助软件MATLAB7．1中神经网络工具箱，将RBF神经网络的理论和算法应用到恒流静压轴承静刚度预测中。经过对物理样机进行相关试验，把测量获得的样本数据用于对RBF神经网络进行训练和测试。结果表明，RBF神经网络能够较准确地预测恒流静压轴承的静刚度。%Through analysis of the characteristics of constant -current hydrostatic bearings for NC rotary table SKZT3500,the theory and arithmetic for the RBF neural network are applied to predict the static stiffness for the bear-ings by using neural network toolbox in MATLAB 7.1.The measured sample data is used to train and test the RBF neural network through experiment of physical prototype.The results show that the RBF neural network can accurately forecast the static stiffness of constant-current hydrostatic bearings.
陈龙宪
2012-01-01
As the RBF network can approach any model with any precision,we put forward an adaptive Controller of Robot Based on RBF Network with uncertainty of model approximation.The RBF network can greatly accelerate the learning speed and avoid local minima problems,suitable for real-time control requirements.The simulation results show that the control algorithm has strong robustness and superior tracking capability.%利用RBF网络能以任意精度逼近任意的连续函数的特点,设计一种基于模型不确定逼近RBF网络机器人的自适应控制器。采用RBF网络可以大大加快学习速度,并避免局部极小问题,适合于实时控制要求。仿真结果表明,该控制算法具有较强的鲁棒性和较好的跟踪性。
张爱科; 符保龙; 李辉
2012-01-01
Web文本分类是数据挖掘研究的一个热点问题.针对文本向量维度过高的特点,提出一种改进的模糊聚类RBF网络集成的文本分类方法,该方法利用模糊C均值聚类算法对文本特征向量进行简化、抽取,引入自适应遗传算法优化RBF神经网络的权值,构建RBF网络集成模型对文本进行分类.实验结果表明,该方法具有更高的分类效率和正确率.%Web text classification is a hot issue in the research on data mining. In view of the characteristics of high dimension text vector, the paper proposes an improved text classification method of fuzzy cluster RBF network integration. The method uses fuzzy c-means clustering algorithm to simplify and extract the text eigenvector, introduces adaptive genetic algorithm for optimization of RBF Neural network weights, and builds a RBF network model for text classification. Experimental results show that the method possesses a higher classification efficiency and accuracy.
Method of generating features optimal to a dataset and classifier
Bruillard, Paul J.; Gosink, Luke J.; Jarman, Kenneth D.
2016-10-18
A method of generating features optimal to a particular dataset and classifier is disclosed. A dataset of messages is inputted and a classifier is selected. An algebra of features is encoded. Computable features that are capable of describing the dataset from the algebra of features are selected. Irredundant features that are optimal for the classifier and the dataset are selected.
武玉英; 李豪; 蒋国瑞
2015-01-01
In order to improve the self-learning ability of traditional negotiation,this paper integrated multi-agent intelligent technology ,designed the negotiation framework based on the blackboard model,constructed the five-elements negotiation model,adopted the negotiation strategy based on Q-reinforcement learning,proposed a negotiation strategy;then it optimized the negotiation strategy by the RBF neural network,predicted the information of opponent for adjusting the concession extent. At last,it verifies the feasibility and validity of the algorithm through a sample application.When comparing to the un-opti-mized Q-reinforcement learning,it can enhance the learning ability of the negotiation agents,reduce the negotiation time,and improve the efficiency of resolving conflicts.%为提高传统协商自学习能力，利用多 agent 智能技术，建立基于黑板模型的协商框架，构建五元组协商模型，采取 Q-强化学习算法，给出一种协商策略；使用 RBF 神经网络进一步优化协商策略，预测对手信息并调整让步幅度。通过算例验证该方法的可行性和有效性，通过与未改进的 Q-强化学习算法对比，该方法可增强协商agent 的自学习能力，缩短协商时间，提高冲突消解效率。
Using RBF to Enable Circuit Emulation Service over Internet%采用RBF来支撑互联网络上的电路模拟服务
金涬; 张斌; 赵阳; 王庆波; 陈滢
2009-01-01
Circuit Emulation Service(CES)aims to enable packet switched networks to provide guaranteed services with comparable qualities of circuit switched networks.Our paper addresses the key issue of QoS of CES flows over Internet.Enlightened by the time division idea popularly used in circuit switched networks,we propose a time division based control mechanism to provide guaranteed QoS for the constant-rate CES flows.The control mechanism is able to estimate the arrival times of the coming packets in CES flows,and reserve the time slots for them.ACCOrdingly.it enables the packets to consume the reserved time slots of their own,so the CES flows are guaranteed to be processed.Refreshing Bloom Filter(RBF),an efficient data representation structure,is proposed to support the time division control mechanism.It consists of multiple bloom filters,and can efficiently record the arrival time slots of millions of packets.The proposed control system model could be a practical tool to support Circuit Emulation Services over Intemet.
Babita Majhi
2014-09-01
Full Text Available This paper develops and assesses the performance of a hybrid prediction model using a radial basis function neural network and non-dominated sorting multiobjective genetic algorithm-II (NSGA-II for various stock market forecasts. The proposed technique simultaneously optimizes two mutually conflicting objectives: the structure (the number of centers in the hidden layer and the output mean square error (MSE of the model. The best compromised non-dominated solution-based model was determined from the optimal Pareto front using fuzzy set theory. The performances of this model were evaluated in terms of four different measures using Standard and Poor 500 (S&P500 and Dow Jones Industrial Average (DJIA stock data. The results of the simulation of the new model demonstrate a prediction performance superior to that of the conventional radial basis function (RBF-based forecasting model in terms of the mean average percentage error (MAPE, directional accuracy (DA, Thelis’ U and average relative variance (ARV values.
Assareh, Ehsanolah; Poultangari, Iman [Dezful Branch, Islamic Azad University, Dezful (Iran, Islamic Republic of); Tandis, Emad [Mechanical Engineering Department, University of Jundi Shapor, Dezful (Iran, Islamic Republic of); Nedael, Mojtaba [Dept. of Energy Engineering, Graduate School of the Environment and Energy, Science and Research Branch, Islamic Azad University, Tehran (Iran, Islamic Republic of)
2016-10-15
Enhancing the energy production from wind power in low-wind areas has always been a fundamental subject of research in the field of wind energy industry. In the first phase of this research, an initial investigation was performed to evaluate the potential of wind in south west of Iran. The initial results indicate that the wind potential in the studied location is not sufficient enough and therefore the investigated region is identified as a low wind speed area. In the second part of this study, an advanced optimization model was presented to regulate the torque in the wind generators. For this primary purpose, the torque of wind turbine is adjusted using a Proportional and integral (PI) control system so that at lower speeds of the wind, the power generated by generator is enhanced significantly. The proposed model uses the RBF neural network to adjust the net obtained gains of the PI controller for the purpose of acquiring the utmost electricity which is produced through the generator. Furthermore, in order to edify and instruct the neural network, the optimal data set is obtained by a Hybrid genetic algorithm along with a gravitational search algorithm (HGA-GSA). The proposed method is evaluated by using a 5MW wind turbine manufactured by National Renewable Energy Laboratory (NREL). Final results of this study are indicative of the satisfactory and successful performance of the proposed investigated model.
RBF neural network prediction algorithm for zero speed parking of elevator%电梯零速停靠的RBF神经网络预测算法
丁宝; 唐海燕; 丁艳虹; 齐维贵
2009-01-01
针对电梯运行过程中存在爬行距离的问题,提出了基于RBF(Radial Basis Function)神经网络的爬行距离预测模型.将预测的爬行距离增加到电梯速度曲线的匀速段,实现减小或消除爬行距离的目的,从而实现电梯的零速停靠.从电梯运行现场采集大量的原始数据,建立RBF神经网络预测模型,与BP(BackPropagation)预测方法进行仿真比较,结果表明RBF神经网络具有更好的预测效果.给出了应用零速停靠RBF预测算法前后电梯运行的速度曲线,爬行距离减小或消除,电梯的运行时间变短,实现了节能.
Equalizer Based on RBF Neural Network and RLS Algorithm%基于RBF神经网络与RLS算法的均衡器
吕志胜; 赖惠成
2009-01-01
将径向基函数神经网络与横向均衡器相结合,采用递推最小二乘算法更新权值.将最小二乘误差作为代价函数以及与误差相关的变步长,使输出误差较传统的神经网络均衡器进一步减小,收敛速度得到提高.仿真结果表明,该均衡器对线性信道和非线性信道都表现出较好的性能,在较严重的非线性情况下其优越性更明显.%This paper combines Radial Base Function(RBF) neural network and landscape filter, uses Recursive Least Square(RLS) algorithm to update the weight and uses variable steps associated with errors, the output error and the convergence speed are both improved. Simulations results show that the new equalizer has better performance, whether it is in linear or nonlinear. In more serious cases, its advantages are much more obvious.
基于RBF神经网络的老年痴呆症智能诊断研究%Study on Intelligent Diagnosis of Senile Dementia Based on RBF Neural Network
张会敏; 叶明全; 罗永钱; 孟婷玮; 陈玥珠
2015-01-01
In order to verify single RBF neural network is more suitable for the predictive diagnosis of senile dementia, through the simulation experiment, a single BP neural network, a single RBF neural network, a genetic algorithm to optimize BP neural network and a genetic algorithm to optimize RBF neural network are used to predict senile dementia, establishing of these four kinds of network model, then analyzing and comparing the forecasted results of these four kinds of network model. The simulation experiments were carried out on the platform of Matlab software, the results show that: in the predictive diagnosis of senile dementia, the single RBF neural network predictive results is higher than the single BP neural network,and the modeling time is shorter. Furthermore, the prediction results of the single RBF neural network is as the same as the genetic algorithm to optimize BP neural network, but the single RBF neural network model is relatively simple, and the prediction results are more stable. Therefore, diagnosis and prediction of the single RBF neural network is more suitable for senile dementia, and this conclusion can be used as a theoretical guide to the actual application.%为验证单RBF神经网络更适用于老年痴呆症的预测诊断，通过仿真实验将单BP神经网络、单RBF神经网络、遗传算法优化BP神经网络及遗传算法优化RBF神经网络分别应用于老年痴呆症的预测诊断，建立这四种网络模型，并对四种网络模型的预测结果进行分析比较。仿真实验在Matlab软件平台上进行。结果表明：在老年痴呆症的预测诊断中，单RBF神经网络比单BP神经网络预测结果更好，建模时间更短。此外，单RBF神经网络与遗传算法优化的BP神经网络预测结果相同，但单RBF神经网络建模较为简单，预测结果更为稳定。而遗传算法对RBF神经网络优化作用不明显。因此，单RBF神经网络更适用于老年痴呆症的预测诊
Confidence-Based Feature Acquisition
Wagstaff, Kiri L.; desJardins, Marie; MacGlashan, James
2010-01-01
Confidence-based Feature Acquisition (CFA) is a novel, supervised learning method for acquiring missing feature values when there is missing data at both training (learning) and test (deployment) time. To train a machine learning classifier, data is encoded with a series of input features describing each item. In some applications, the training data may have missing values for some of the features, which can be acquired at a given cost. A relevant JPL example is that of the Mars rover exploration in which the features are obtained from a variety of different instruments, with different power consumption and integration time costs. The challenge is to decide which features will lead to increased classification performance and are therefore worth acquiring (paying the cost). To solve this problem, CFA, which is made up of two algorithms (CFA-train and CFA-predict), has been designed to greedily minimize total acquisition cost (during training and testing) while aiming for a specific accuracy level (specified as a confidence threshold). With this method, it is assumed that there is a nonempty subset of features that are free; that is, every instance in the data set includes these features initially for zero cost. It is also assumed that the feature acquisition (FA) cost associated with each feature is known in advance, and that the FA cost for a given feature is the same for all instances. Finally, CFA requires that the base-level classifiers produce not only a classification, but also a confidence (or posterior probability).
Repositioning Recitation Input in College English Teaching
Xu, Qing
2009-01-01
This paper tries to discuss how recitation input helps overcome the negative influences on the basis of second language acquisition theory and confirms the important role that recitation input plays in improving college students' oral and written English.
Tunable Versatile High Input Impedance Voltage-Mode Universal Biquadratic Filter Based on DDCCs
Jiun-Wei Horng
2012-12-01
Full Text Available A high input impedance voltage-mode universal biquadratic filter with three input terminals and seven output terminals is presented. The proposed circuit uses three differential difference current conveyors (DDCCs, four resistors and two grounded capacitors. The proposed circuit can realize all the standard filter functions, namely, lowpass, bandpass, highpass, notch and allpass, simultaneously. The proposed circuit offers the features of high input impedance, using only grounded capacitors, and orthogonal controllability of resonance angular frequency and quality factor.
Soft computing based feature selection for environmental sound classification
Shakoor, A.; May, T.M.; Van Schijndel, N.H.
2010-01-01
Environmental sound classification has a wide range of applications,like hearing aids, mobile communication devices, portable media players, and auditory protection devices. Sound classification systemstypically extract features from the input sound. Using too many features increases complexity unne
Soft computing based feature selection for environmental sound classification
Shakoor, A.; May, T.M.; Van Schijndel, N.H.
2010-01-01
Environmental sound classification has a wide range of applications,like hearing aids, mobile communication devices, portable media players, and auditory protection devices. Sound classification systemstypically extract features from the input sound. Using too many features increases complexity unne
Facilitating agricultural input distribution in Uganda - Experiences ...
Mo
The input supply market however, suffered a setback as a result of the ... Ltd. redefined the approach emphasizing a demand driven input market by shifting ... Training of business entrepreneurs in business planning, ... The strategy to increase rural demand for agricultural inputs ..... During season 2004A, the basic fertilizers.
Effects of Auditory Input in Individuation Tasks
Robinson, Christopher W.; Sloutsky, Vladimir M.
2008-01-01
Under many conditions auditory input interferes with visual processing, especially early in development. These interference effects are often more pronounced when the auditory input is unfamiliar than when the auditory input is familiar (e.g. human speech, pre-familiarized sounds, etc.). The current study extends this research by examining how…
7 CFR 3430.607 - Stakeholder input.
2010-01-01
... 7 Agriculture 15 2010-01-01 2010-01-01 false Stakeholder input. 3430.607 Section 3430.607 Agriculture Regulations of the Department of Agriculture (Continued) COOPERATIVE STATE RESEARCH, EDUCATION... § 3430.607 Stakeholder input. CSREES shall seek and obtain stakeholder input through a variety of...
7 CFR 3430.15 - Stakeholder input.
2010-01-01
... 7 Agriculture 15 2010-01-01 2010-01-01 false Stakeholder input. 3430.15 Section 3430.15... Stakeholder input. Section 103(c)(2) of the Agricultural Research, Extension, and Education Reform Act of 1998... RFAs for competitive programs. CSREES will provide instructions for submission of stakeholder input...
7 CFR 3430.907 - Stakeholder input.
2010-01-01
... 7 Agriculture 15 2010-01-01 2010-01-01 false Stakeholder input. 3430.907 Section 3430.907 Agriculture Regulations of the Department of Agriculture (Continued) COOPERATIVE STATE RESEARCH, EDUCATION... Program § 3430.907 Stakeholder input. CSREES shall seek and obtain stakeholder input through a variety...
Functional properties of GABA synaptic inputs onto GABA neurons in monkey prefrontal cortex
D.C. Rotaru (Diana C.); C. Olezene (Cameron); T. Miyamae (Takeaki); N.V. Povysheva (Nadezhda V.); A.V. Zaitsev (Aleksey V.); D.A. Lewis (David A.); G. Gonzalez-Burgos (Guillermo)
2015-01-01
textabstractIn rodent cortex GABA
Input calibration for negative originals
Tuijn, Chris
1995-04-01
One of the major challenges in the prepress environment consists of controlling the electronic color reproduction process such that a perfect match of any original can be realized. Whether this goal can be reached depends on many factors such as the dynamic range of the input device (scanner, camera), the color gamut of the output device (dye sublimation printer, ink-jet printer, offset), the color management software etc. The characterization of the color behavior of the peripheral devices is therefore very important. Photographs and positive transparents reflect the original scene pretty well; for negative originals, however, there is no obvious link to either the original scene or a particular print of the negative under consideration. In this paper, we establish a method to scan negatives and to convert the scanned data to a calibrated RGB space, which is known colorimetrically. This method is based on the reconstruction of the original exposure conditions (i.e., original scene) which generated the negative. Since the characteristics of negative film are quite diverse, a special calibration is required for each combination of scanner and film type.
关海鸥; 杜松怀; 李春兰; 苏娟; 梁英; 武子超; 邵利敏
2013-01-01
针对农村低压电网剩余电流保护与动作技术中，如何检测总泄漏电流中人体触电支路电流的难题，该文利用严格线性相位与任意幅度特性的FIR(finite impulse response)数字滤波技术和具有自适应性与最佳逼近特性的RBF(radial basis function)神经网络有机结合，提出一种基于FIR数字滤波的RBF神经网络作为触电电流信号的检测方法.首先，采用FIR数字滤波器选定合适的窗函数和滤波阶数，对触电试验获得的总泄漏电流及触电电流进行滤波预处理；然后，将预处理后的信号波形作为样本集，选定适合的RBF函数，建立从总泄漏电流中提取触电电流波形的3层RBF神经网络模型.仿真试验结果表明：该方法速度快且稳定，检测值与实际值的平均相对误差为3.76%，具有良好的适应性和实用性，对于研制新一代剩余电流保护动作装置具有重要意义.%Residual current protection device (RCD) has been widely used in low-voltage, rural power grids because it plays a very important role in avoiding physical shock, equipment damage, and electrical fires, etc, caused by leakage. At present, a setting value of leakage current can often be used as a key action for RCD. However, the electric shock current signal of the human body cannot be detected, and when unexpected current values close to or more than the setting value emerge, this will lead to the malfunction of RCD. In order to overcome the shortcomings above, we present a new recognition method for electric shock signal using digital filter technology and radial basis neural network. The method has three main stages. First, total leakage current and electric short current has been pre-processed using the finite impulse response digital filtering, which was designed by choosing suitable window functions and filter order. Second, the pre-processed signals are trained to create a three-level radial basis neural network
Application of Grammatical Parsing Technique in Chinese Input
俞士汶
1990-01-01
In Peking University Computer Research Institute(PUCRI) a method of inputting Chinese sentences based on words has been developed.To reduce the troubles in choosing one word out of the others characterized by the same feature,grammatical parsing technique is applied to the method and good results have been achieved.This article describes the outline of the method.the principle of applying grammatical formulas and the branch-cutting algorithm used to speed up the grammatical parsing.
The use of synthetic input sequences in time series modeling
Oliveira, Dair Jose de [Programa de Pos-Graduacao em Engenharia Eletrica, Universidade Federal de Minas Gerais, Av. Antonio Carlos 6627, 31.270-901 Belo Horizonte, MG (Brazil); Letellier, Christophe [CORIA/CNRS UMR 6614, Universite et INSA de Rouen, Av. de l' Universite, BP 12, F-76801 Saint-Etienne du Rouvray cedex (France); Gomes, Murilo E.D. [Programa de Pos-Graduacao em Engenharia Eletrica, Universidade Federal de Minas Gerais, Av. Antonio Carlos 6627, 31.270-901 Belo Horizonte, MG (Brazil); Aguirre, Luis A. [Programa de Pos-Graduacao em Engenharia Eletrica, Universidade Federal de Minas Gerais, Av. Antonio Carlos 6627, 31.270-901 Belo Horizonte, MG (Brazil)], E-mail: aguirre@cpdee.ufmg.br
2008-08-04
In many situations time series models obtained from noise-like data settle to trivial solutions under iteration. This Letter proposes a way of producing a synthetic (dummy) input, that is included to prevent the model from settling down to a trivial solution, while maintaining features of the original signal. Simulated benchmark models and a real time series of RR intervals from an ECG are used to illustrate the procedure.
The use of synthetic input sequences in time series modeling
de Oliveira, Dair José; Letellier, Christophe; Gomes, Murilo E. D.; Aguirre, Luis A.
2008-08-01
In many situations time series models obtained from noise-like data settle to trivial solutions under iteration. This Letter proposes a way of producing a synthetic (dummy) input, that is included to prevent the model from settling down to a trivial solution, while maintaining features of the original signal. Simulated benchmark models and a real time series of RR intervals from an ECG are used to illustrate the procedure.
Learning to represent visual input.
Hinton, Geoffrey E
2010-01-12
One of the central problems in computational neuroscience is to understand how the object-recognition pathway of the cortex learns a deep hierarchy of nonlinear feature detectors. Recent progress in machine learning shows that it is possible to learn deep hierarchies without requiring any labelled data. The feature detectors are learned one layer at a time and the goal of the learning procedure is to form a good generative model of images, not to predict the class of each image. The learning procedure only requires the pairwise correlations between the activations of neuron-like processing units in adjacent layers. The original version of the learning procedure is derived from a quadratic 'energy' function but it can be extended to allow third-order, multiplicative interactions in which neurons gate the pairwise interactions between other neurons. A technique for factoring the third-order interactions leads to a learning module that again has a simple learning rule based on pairwise correlations. This module looks remarkably like modules that have been proposed by both biologists trying to explain the responses of neurons and engineers trying to create systems that can recognize objects.
Turn customer input into innovation.
Ulwick, Anthony W
2002-01-01
It's difficult to find a company these days that doesn't strive to be customer-driven. Too bad, then, that most companies go about the process of listening to customers all wrong--so wrong, in fact, that they undermine innovation and, ultimately, the bottom line. What usually happens is this: Companies ask their customers what they want. Customers offer solutions in the form of products or services. Companies then deliver these tangibles, and customers just don't buy. The reason is simple--customers aren't expert or informed enough to come up with solutions. That's what your R&D team is for. Rather, customers should be asked only for outcomes--what they want a new product or service to do for them. The form the solutions take should be up to you, and you alone. Using Cordis Corporation as an example, this article describes, in fine detail, a series of effective steps for capturing, analyzing, and utilizing customer input. First come indepth interviews, in which a moderator works with customers to deconstruct a process or activity in order to unearth "desired outcomes." Addressing participants' comments one at a time, the moderator rephrases them to be both unambiguous and measurable. Once the interviews are complete, researchers then compile a comprehensive list of outcomes that participants rank in order of importance and degree to which they are satisfied by existing products. Finally, using a simple mathematical formula called the "opportunity calculation," researchers can learn the relative attractiveness of key opportunity areas. These data can be used to uncover opportunities for product development, to properly segment markets, and to conduct competitive analysis.
An Optimal SVM with Feature Selection Using Multiobjective PSO
Iman Behravan
2016-01-01
Full Text Available Support vector machine is a classifier, based on the structured risk minimization principle. The performance of the SVM depends on different parameters such as penalty factor, C, and the kernel factor, σ. Also choosing an appropriate kernel function can improve the recognition score and lower the amount of computation. Furthermore, selecting the useful features among several features in dataset not only increases the performance of the SVM, but also reduces the computational time and complexity. So this is an optimization problem which can be solved by heuristic algorithm. In some cases besides the recognition score, the reliability of the classifier’s output is important. So in such cases a multiobjective optimization algorithm is needed. In this paper we have got the MOPSO algorithm to optimize the parameters of the SVM, choose appropriate kernel function, and select the best feature subset simultaneously in order to optimize the recognition score and the reliability of the SVM concurrently. Nine different datasets, from UCI machine learning repository, are used to evaluate the power and the effectiveness of the proposed method (MOPSO-SVM. The results of the proposed method are compared to those which are achieved by single SVM, RBF, and MLP neural networks.
A new approach for detecting local features
Nguyen, Phuong Giang; Andersen, Hans Jørgen
2010-01-01
Local features up to now are often mentioned in the meaning of interest points. A patch around each point is formed to compute descriptors or feature vectors. Therefore, in order to satisfy different invariant imaging conditions such as scales and viewpoints, an input image is often represented i...
Adrian Ion-Mărgineanu
2017-07-01
Full Text Available Purpose: The purpose of this study is classifying multiple sclerosis (MS patients in the four clinical forms as defined by the McDonald criteria using machine learning algorithms trained on clinical data combined with lesion loads and magnetic resonance metabolic features.Materials and Methods: Eighty-seven MS patients [12 Clinically Isolated Syndrome (CIS, 30 Relapse Remitting (RR, 17 Primary Progressive (PP, and 28 Secondary Progressive (SP] and 18 healthy controls were included in this study. Longitudinal data available for each MS patient included clinical (e.g., age, disease duration, Expanded Disability Status Scale, conventional magnetic resonance imaging and spectroscopic imaging. We extract N-acetyl-aspartate (NAA, Choline (Cho, and Creatine (Cre concentrations, and we compute three features for each spectroscopic grid by averaging metabolite ratios (NAA/Cho, NAA/Cre, Cho/Cre over good quality voxels. We built linear mixed-effects models to test for statistically significant differences between MS forms. We test nine binary classification tasks on clinical data, lesion loads, and metabolic features, using a leave-one-patient-out cross-validation method based on 100 random patient-based bootstrap selections. We compute F1-scores and BAR values after tuning Linear Discriminant Analysis (LDA, Support Vector Machines with gaussian kernel (SVM-rbf, and Random Forests.Results: Statistically significant differences were found between the disease starting points of each MS form using four different response variables: Lesion Load, NAA/Cre, NAA/Cho, and Cho/Cre ratios. Training SVM-rbf on clinical and lesion loads yields F1-scores of 71–72% for CIS vs. RR and CIS vs. RR+SP, respectively. For RR vs. PP we obtained good classification results (maximum F1-score of 85% after training LDA on clinical and metabolic features, while for RR vs. SP we obtained slightly higher classification results (maximum F1-score of 87% after training LDA and SVM-rbf
Live facial feature extraction
ZHAO JieYu
2008-01-01
Precise facial feature extraction is essential to the high-level face recognition and expression analysis. This paper presents a novel method for the real-time geomet-ric facial feature extraction from live video. In this paper, the input image is viewed as a weighted graph. The segmentation of the pixels corresponding to the edges of facial components of the mouth, eyes, brows, and nose is implemented by means of random walks on the weighted graph. The graph has an 8-connected lattice structure and the weight value associated with each edge reflects the likelihood that a random walker will cross that edge. The random walks simulate an anisot-ropic diffusion process that filters out the noise while preserving the facial expres-sion pixels. The seeds for the segmentation are obtained from a color and motion detector. The segmented facial pixels are represented with linked lists in the origi-nal geometric form and grouped into different parts corresponding to facial com-ponents. For the convenience of implementing high-level vision, the geometric description of facial component pixels is further decomposed into shape and reg-istration information. Shape is defined as the geometric information that is invari-ant under the registration transformation, such as translation, rotation, and iso-tropic scale. Statistical shape analysis is carried out to capture global facial fea-tures where the Procrustes shape distance measure is adopted. A Bayesian ap-proach is used to incorporate high-level prior knowledge of face structure. Ex-perimental results show that the proposed method is capable of real-time extraction of precise geometric facial features from live video. The feature extraction is robust against the illumination changes, scale variation, head rotations, and hand inter-ference.
Image processing tool for automatic feature recognition and quantification
Chen, Xing; Stoddard, Ryan J.
2017-05-02
A system for defining structures within an image is described. The system includes reading of an input file, preprocessing the input file while preserving metadata such as scale information and then detecting features of the input file. In one version the detection first uses an edge detector followed by identification of features using a Hough transform. The output of the process is identified elements within the image.
Regional input-output models and the treatment of imports in the European System of Accounts
Kronenberg, Tobias
2011-01-01
Input-output models are often used in regional science due to their versatility and their ability to capture many of the distinguishing features of a regional economy. Input-output tables are available for all EU member countries, but they are hard to find at the regional level, since many regional governments lack the resources or the will to produce reliable, survey-based regional input-output tables. Therefore, in many cases researchers adopt nonsurvey techniques to derive regional input-o...
Evaluation of feature detection algorithms for structure from motion
Govender, N
2009-11-01
Full Text Available such as Harris corner detectors and feature descriptors such as SIFT (Scale Invariant Feature Transform) and SURF (Speeded Up Robust Features) given a set of input images. This paper implements state-of-the art feature detection algorithms and evaluates...
Elizabeth Ritter
2015-01-01
Full Text Available Algonquian languages are famous for their animacy-based grammatical properties—an animacy based noun classification system and direct/inverse system which gives rise to animacy hierarchy effects in the determination of verb agreement. In this paper I provide new evidence for the proposal that the distinctive properties of these languages is due to the use of participant-based features, rather than spatio-temporal ones, for both nominal and verbal functional categories (Ritter & Wiltschko 2009, 2014. Building on Wiltschko (2012, I develop a formal treatment of the Blackfoot aspectual system that assumes a category Inner Aspect (cf. MacDonald 2008, Travis 1991, 2010. Focusing on lexical aspect in Blackfoot, I demonstrate that the classification of both nouns (Seinsarten and verbs (Aktionsarten is based on animacy, rather than boundedness, resulting in a strikingly different aspectual system for both categories.
Input estimation from measured structural response
Harvey, Dustin [Los Alamos National Laboratory; Cross, Elizabeth [Los Alamos National Laboratory; Silva, Ramon A [Los Alamos National Laboratory; Farrar, Charles R [Los Alamos National Laboratory; Bement, Matt [Los Alamos National Laboratory
2009-01-01
This report will focus on the estimation of unmeasured dynamic inputs to a structure given a numerical model of the structure and measured response acquired at discrete locations. While the estimation of inputs has not received as much attention historically as state estimation, there are many applications where an improved understanding of the immeasurable input to a structure is vital (e.g. validating temporally varying and spatially-varying load models for large structures such as buildings and ships). In this paper, the introduction contains a brief summary of previous input estimation studies. Next, an adjoint-based optimization method is used to estimate dynamic inputs to two experimental structures. The technique is evaluated in simulation and with experimental data both on a cantilever beam and on a three-story frame structure. The performance and limitations of the adjoint-based input estimation technique are discussed.
Input Method "Five Strokes": Advantages and Problems
Mateja PETROVČIČ
2014-03-01
Since the Five Stroke input method is easily accessible, simple to master and is not pronunciation-based, we would expect that the students will use it to input unknown characters. The survey comprises students of Japanology and Sinology at Department of Asian and African Studies, takes in consideration the grade of the respondent and therefore his/her knowledge of characters. This paper also discusses the impact of typeface to the accuracy of the input.
卫晓娟; 丁旺才; 李宁洲; 郭文志
2016-01-01
为解决神经网络结构及参数的优化选择问题，以提高机车齿轮箱故障诊断的精度，提出一种基于引力搜索RB F神经网络的机车齿轮箱智能故障诊断方法。基于高斯RB F神经网络建立机车齿轮箱故障诊断模型，采用减聚类算法确定RB F神经网络结构，并结合混沌优化策略及人工蜂群搜索算子提出自适应混合引力搜索算法对故障诊断模型进行优化求解，避免了参数选择的盲目性。采用国际标准测试数据集对该方法进行分类性能测试，结果表明其分类精度明显优于经GA算法、SPSO算法、QPSO算法和GSA算法优化的RBF神经网络。将该方法应用于机车齿轮箱故障的诊断，应用实例验证了该方法的有效性。%In order to solve the issue of the determination of neural network structure and the optimization of neural network parameters to improve the accuracy of fault diagnosis of locomotive gearbox ,an intelligent fault diagnosis method based on the gravitational search algorithm and RBF neural network was proposed . When the locomotive gearbox fault diagnosis model was established based on Gaussian RBF neural network , subtractive clustering algorithm was used to determine the structure of RBF neural network .By reference to the artificial bee colony search operator and chaos optimization strategy , an adaptive hybrid gravitational search algorithm was proposed and applied to solve and optimize the fault diagnosis model , to avoid the blindness of parameter selection . Results of the classification performance test on the proposed method using UCI testing data sets showed that the classification accuracy of the proposed method was significantly better than the RBF neural net‐work optimized by GA algorithm ,SPSO algorithm ,QPSO algorithm and GSA algorithm . The application of the proposed method in fault diagnosis of locomotive gearbox demonstrated the effectiveness of this method .
吴一晓; 杨然; 李占军; 何浩强; 胡红丽
2014-01-01
Objective:To design a knee Osteoarthritis classifier of magnetic resonance T2 map data, which used for OA disease classification. Methods:Collected 46 cases of knee image with a total of 1380 data by magnetic resonance imaging (MRI) T2 mapping technique, and extracted T2 value data of 10 Asian region based on articular cartilage whole organ magnetic resonance imaging score (WORMS) partition. Then took the T2 value data as the characteristic quantity by data mining, and structured radial basis function (RBF) neural network classifier, combined with the clinical diagnosis to classify and recognize the data of collected sample. Results:The study finally found that RBF classifier reflected 75% of recognition accuracy rate, and it showed good effect of OA data classification. Conclusion:The knee osteoarthritis RBF neural network classifier based on direct determination method can get the optimal weights, right center and variance by simple steps, without any iteration. We suggest that it is a classifier fit to OA disease.%目的：设计一种膝关节骨性关节炎(OA)磁共振T2 map数据分类器，用于OA疾病分类。方法：通过磁共振成像(MRI)T2 mapping技术，采集46例膝关节MRI图像共计1380个数据，按膝关节软骨全器官磁共振成像评分(WORMS)分区方法提取10个亚区的T2值数据，以T2值数据为特征量进行数据挖掘，建立径向基函数(RBF)神经网络分类器，结合临床诊断结果实行对采集样本数据分类识别。结果：RBF分类器对于膝关节T2 map数据最终识别准确率为75%，体现了良好的OA数据分类效果。结论：基于直接确定法的RBF神经网络构造的膝关节OA分类器无需任何迭代，通过简单步骤就得到最优权值、合适的中心以及方差，适合作为OA的疾病分类器。
Feature-Weighted Linear Stacking
Sill, Joseph; Mackey, Lester; Lin, David
2009-01-01
Ensemble methods, such as stacking, are designed to boost predictive accuracy by blending the predictions of multiple machine learning models. Recent work has shown that the use of meta-features, additional inputs describing each example in a dataset, can boost the performance of ensemble methods, but the greatest reported gains have come from nonlinear procedures requiring significant tuning and training time. Here, we present a linear technique, Feature-Weighted Linear Stacking (FWLS), that incorporates meta-features for improved accuracy while retaining the well-known virtues of linear regression regarding speed, stability, and interpretability. FWLS combines model predictions linearly using coefficients that are themselves linear functions of meta-features. This technique was a key facet of the solution of the second place team in the recently concluded Netflix Prize competition. Significant increases in accuracy over standard linear stacking is demonstrated on the Netflix Prize collaborative filtering da...
Input Response of Neural Network Model with Lognormally Distributed Synaptic Weights
Nagano, Yoshihiro; Karakida, Ryo; Watanabe, Norifumi; Aoyama, Atsushi; Okada, Masato
2016-07-01
Neural assemblies in the cortical microcircuit can sustain irregular spiking activity without external inputs. On the other hand, neurons exhibit rich evoked activities driven by sensory stimulus, and both activities are reported to contribute to cognitive functions. We studied the external input response of the neural network model with lognormally distributed synaptic weights. We show that the model can achieve irregular spontaneous activity and population oscillation depending on the presence of external input. The firing rate distribution was maintained for the external input, and the order of firing rates in evoked activity reflected that in spontaneous activity. Moreover, there were bistable regions in the inhibitory input parameter space. The bimodal membrane potential distribution, which is a characteristic feature of the up-down state, was obtained under such conditions. From these results, we can conclude that the model displays various evoked activities due to the external input and is biologically plausible.
Bilinearity in spatiotemporal integration of synaptic inputs.
Songting Li
2014-12-01
Full Text Available Neurons process information via integration of synaptic inputs from dendrites. Many experimental results demonstrate dendritic integration could be highly nonlinear, yet few theoretical analyses have been performed to obtain a precise quantitative characterization analytically. Based on asymptotic analysis of a two-compartment passive cable model, given a pair of time-dependent synaptic conductance inputs, we derive a bilinear spatiotemporal dendritic integration rule. The summed somatic potential can be well approximated by the linear summation of the two postsynaptic potentials elicited separately, plus a third additional bilinear term proportional to their product with a proportionality coefficient [Formula: see text]. The rule is valid for a pair of synaptic inputs of all types, including excitation-inhibition, excitation-excitation, and inhibition-inhibition. In addition, the rule is valid during the whole dendritic integration process for a pair of synaptic inputs with arbitrary input time differences and input locations. The coefficient [Formula: see text] is demonstrated to be nearly independent of the input strengths but is dependent on input times and input locations. This rule is then verified through simulation of a realistic pyramidal neuron model and in electrophysiological experiments of rat hippocampal CA1 neurons. The rule is further generalized to describe the spatiotemporal dendritic integration of multiple excitatory and inhibitory synaptic inputs. The integration of multiple inputs can be decomposed into the sum of all possible pairwise integration, where each paired integration obeys the bilinear rule. This decomposition leads to a graph representation of dendritic integration, which can be viewed as functionally sparse.
Phonological Feature Re-Assembly and the Importance of Phonetic Cues
Archibald, John
2009-01-01
It is argued that new phonological features can be acquired in second languages, but that both feature acquisition and feature re-assembly are affected by the robustness of phonetic cues in the input.
Computing Functions by Approximating the Input
Goldberg, Mayer
2012-01-01
In computing real-valued functions, it is ordinarily assumed that the input to the function is known, and it is the output that we need to approximate. In this work, we take the opposite approach: we show how to compute the values of some transcendental functions by approximating the input to these functions, and obtaining exact answers for their…
Wave energy input into the Ekman layer
无
2008-01-01
This paper is concerned with the wave energy input into the Ekman layer, based on 3 observational facts that surface waves could significantly affect the profile of the Ekman layer. Under the assumption of constant vertical diffusivity, the analytical form of wave energy input into the Ekman layer is derived. Analysis of the energy balance shows that the energy input to the Ekman layer through the wind stress and the interaction of the Stokes-drift with planetary vorticity can be divided into two kinds. One is the wind energy input, and the other is the wave energy input which is dependent on wind speed, wave characteristics and the wind direction relative to the wave direction. Estimates of wave energy input show that wave energy input can be up to 10% in high-latitude and high-wind speed areas and higher than 20% in the Antarctic Circumpolar Current, compared with the wind energy input into the classical Ekman layer. Results of this paper are of significance to the study of wave-induced large scale effects.
Input-dependent wave attenuation in a critically-balanced model of cortex.
Xiao-Hu Yan
Full Text Available A number of studies have suggested that many properties of brain activity can be understood in terms of critical systems. However it is still not known how the long-range susceptibilities characteristic of criticality arise in the living brain from its local connectivity structures. Here we prove that a dynamically critically-poised model of cortex acquires an infinitely-long ranged susceptibility in the absence of input. When an input is presented, the susceptibility attenuates exponentially as a function of distance, with an increasing spatial attenuation constant (i.e., decreasing range the larger the input. This is in direct agreement with recent results that show that waves of local field potential activity evoked by single spikes in primary visual cortex of cat and macaque attenuate with a characteristic length that also increases with decreasing contrast of the visual stimulus. A susceptibility that changes spatial range with input strength can be thought to implement an input-dependent spatial integration: when the input is large, no additional evidence is needed in addition to the local input; when the input is weak, evidence needs to be integrated over a larger spatial domain to achieve a decision. Such input-strength-dependent strategies have been demonstrated in visual processing. Our results suggest that input-strength dependent spatial integration may be a natural feature of a critically-balanced cortical network.
苗青; 曹广益; 朱新坚
2006-01-01
The temperature models of anode and cathode of direct methanol fuel cell (DMFC) stack were established by using radial basis function (RBF) neural networks identification technique to deal with the modeling and control problem of DMFC stack. An adaptive fuzzy neural networks temperature controller was designed based on the identification models established, and parameters of the controller were regulated by novel back propagation (BP) algorithm. Simulation results show that the RBF neural networks identification modeling method is correct, effective and the models established have good accuracy. Moreover, performance of the adaptive fuzzy neural networks temperature controller designed is superior.
Statistical identification of effective input variables. [SCREEN
Vaurio, J.K.
1982-09-01
A statistical sensitivity analysis procedure has been developed for ranking the input data of large computer codes in the order of sensitivity-importance. The method is economical for large codes with many input variables, since it uses a relatively small number of computer runs. No prior judgemental elimination of input variables is needed. The sceening method is based on stagewise correlation and extensive regression analysis of output values calculated with selected input value combinations. The regression process deals with multivariate nonlinear functions, and statistical tests are also available for identifying input variables that contribute to threshold effects, i.e., discontinuities in the output variables. A computer code SCREEN has been developed for implementing the screening techniques. The efficiency has been demonstrated by several examples and applied to a fast reactor safety analysis code (Venus-II). However, the methods and the coding are general and not limited to such applications.
Atmospheric Nitrogen input to the Kattegat
Asman, W.A.H.; Hertel, O.; Berkowicz, R.
1995-01-01
An overview is given of the processes involved in the atmospheric deposition of nitrogen compounds. These processes are incorporated in an atmospheric transport model that is used to calculate the nitrogen input to the Kattegat, the sea area between Denmark and Sweden. The model results show...... that the total atmospheric nitrogen input to the Kattegat is approximately 960 kg N km(-2) yr(-1). The nitrogen input to the Kattegat is dominated by the wet depositions of NHx (42%) and NOy (30%). The contribution from the dry deposition of NHx is 17% and that of the dry deposition of NOy is 11......%. The contribution of the atmospheric input of nitrogen to the Kattegat is about 30% of the total input including the net transport from other sea areas, runoff etc....
Application of RBF neural network algorithm in dynamic weighing%RBF神经网络算法在动态称重中的应用∗
陈超波; 杨楠
2016-01-01
In this paper,focus on the complexity of weighing data in the dynamic detection system of the highway,the different weighing data processing methods to make a comparison,and proposed the use of RBF neural network to deal with the dynamic weighing data.Firstly review introduced the whole vehicle dynamic system,after the radial basis function network topology and the centers of the radial basis function selection are introduced.Finally the test-bed to build a testing platform,through experiments with a two axle vehicle with different speed through the test stand,the dynamic parameters acquisition.Finally,using the data collected,using MATLAB to simulate,verify the radial basis function network to the dynamic weighing data processing show good speed and accuracy.%针对高速公路动态检测系统中称重数据的复杂性，将不同的称重数据处理办法做出对比，并提出利用 RBF神经网络对动态称重数据进行处理。文章首先综述性的介绍了车辆动态系统整体构成，之后对径向基函数网络的拓扑结构以及径向基函数中心的选取进行了介绍，最后利以试验台搭建检测平台，通过实验用两轴小车进行以不同的速度通过试验台，采集其动态参数。最后利用采集到的数据，用 MATLAB 进行仿真，验证了径向基函数网络对动态称重数据的处理表现出良好的速度与精度。
Input impedance characteristics of microstrip structures
A. I. Nazarko
2015-06-01
Full Text Available Introduction. Electromagnetic crystals (EC and EC-inhomogeneities are one of the main directions of microstrip devices development. In the article the input impedance characteristics of EC- and traditional microstrip inhomogeneities and filter based on EC-inhomogeneities are investigated. Transmission coefficient characteristics. Transmission coefficient characteristics of low impedance EC- and traditional inhomogeneities are considered. Characteristics are calculated in the software package Microwave Studio. It is shown that the efficiency of EC-inhomogeneity is much higher. Input impedance characteristics of low impedance inhomogeneities. Dependences of input impedance active and reactive parts of EC- and traditional inhomogeneities are given. Dependences of the active part illustrate significant low impedance transformation of nominal impedance. The conditions of impedance matching of structure and input medium are set. Input impedance characteristics of high impedance inhomogeneities. Input impedance characteristics of high impedance EC- and traditional inhomogeneities are considered. It was shown that the band of transformation by high impedance inhomogeneities is much narrower than one by low impedance inhomogeneities. Characteristics of the reflection coefficient of inhomogeneities are presented. Input impedance characteristics of narrowband filter. The structure of narrowband filter based on the scheme of Fabry-Perot resonator is presented. The structure of the filter is fulfilled by high impedance EC-inhomogeneities as a reflectors. Experimental and theoretical amplitude-frequency characteristics of the filter are presented. Input impedance characteristics of the filter are shown. Conclusions. Input impedance characteristics of the structure allow to analyse its wave properties, especially resonant. EC-inhomogeneity compared with traditional microstrip provide substantially more significant transformation of the the input impedance.
Self-adaptive RBF Neural Network Control Theory of Switched Reluctance Motor%开关磁阻电机自适应RBF神经网络控制方法研究
蔡益春; 王立标
2012-01-01
Switched reluctance motor doubly salient structure and magnetic circuit of the motor flux is highly saturated highly nonlinear, leading to classical PID control can not get higher control precision. Designing of feed-forward and feedback controller based on self-adaptive RBF neural network for SRM, the simulation results show that, this method can improve the precision of motor speed, torque pulsation, thereby optimizing the motor operat-ing performance.%开关磁阻电机(switched reluctance motor,SRM)双凸极结构和磁路高度饱和使得电机磁链呈高度非线性,导致经典PID控制不能得到较高的控制精度.设计了基于自适应RBF(radial basis function)神经网络的SRM前馈+反馈控制器,对电机实行自适应控制.仿真结果表明,该方法能提高电机转速精度,降低转矩脉动,从而优化电机的运行性能.
张高扬; 翟成功; 刁永辉
2013-01-01
本文在纬编摇粒绒织物的保暖性能测试中，利用RBF人工神经网络的智能鉴别和预测功能，在实验测试中进行实验数据的预测、鉴别和修订，排除错误的数据；利用RBF人工神经网络强大的数据处理能力，来进行数据分析，保证数据处理的准确性，以便得到更加客观真实的结论。%This paper is designed to use the intelligent data forecasted and distinguished function of RBF artificial neural network in the thermal test of the weft polar fleece, in order to modify the test data and eliminate mistakes. The test data is analyzed by using RBF artificial neural network to ensure the data pro-cessing veracity and the impersonal conclusion.
Dehghan, Mehdi; Mohammadi, Vahid
2017-08-01
In this research, we investigate the numerical solution of nonlinear Schrödinger equations in two and three dimensions. The numerical meshless method which will be used here is RBF-FD technique. The main advantage of this method is the approximation of the required derivatives based on finite difference technique at each local-support domain as Ωi. At each Ωi, we require to solve a small linear system of algebraic equations with a conditionally positive definite matrix of order 1 (interpolation matrix). This scheme is efficient and its computational cost is same as the moving least squares (MLS) approximation. A challengeable issue is choosing suitable shape parameter for interpolation matrix in this way. In order to overcome this matter, an algorithm which was established by Sarra (2012), will be applied. This algorithm computes the condition number of the local interpolation matrix using the singular value decomposition (SVD) for obtaining the smallest and largest singular values of that matrix. Moreover, an explicit method based on Runge-Kutta formula of fourth-order accuracy will be applied for approximating the time variable. It also decreases the computational costs at each time step since we will not solve a nonlinear system. On the other hand, to compare RBF-FD method with another meshless technique, the moving kriging least squares (MKLS) approximation is considered for the studied model. Our results demonstrate the ability of the present approach for solving the applicable model which is investigated in the current research work.
张殷钦; 刘俊民; 郝健
2011-01-01
【目的】建立地下水位预测的正则化RBF网络模型,为区域地下水资源的利用、规划和管理提供决策依据。【方法】以MATLAB7.0为平台,用函数newrb创建正则化RBF网络模型,基于宝鸡峡灌区B210号观测井1983-2009年的地下水位埋深资料,对网络模型进行训练后再用测试集检验,分别绘制训练集与测试集的拟合曲线,同时计算实测值与预测值间的相对误差（RE）、平均绝对偏差（MAD）和均方误差（MSE）,并将其与BP网络模型的相应值进行对比。【结果】正则化RBF网络模型和BP网络模型的相对误差均小于5%,平均绝对偏差分别为0.53和0.85,均方误差分别为0.54和1.15,相比之下,正则化RBF网络模型的预测精度更高。【结论】训练样本和测试样本的合理选取为时间序列的拟合扩展了思路,良好的泛化能力使正则化RBF网络模型在区域地下水位预测中具有一定的可行性。%【Objective】 Establishing regularized RBF network model for groundwater level prediction can provide strategic decision for groundwater use,planning and management.【Method】 Regularized RBF network model was built employing newrb function in MATLAB7.0 for well B210 in Baojixia irrigation area based on the groundwater level depth data from 1983 to 2009.The training sets and testing sets were used to train and test the network respectively.Corresponding fitting curve was plotted as well.Meanwhile,relative error（RE）,mean absolute deviation（MAD） and mean-square error（MSE） between predicted and measured values were all calculated and the comparison was addressed with BP network model.【Result】 RE of both regularized RBF and BP network model is less than 5%,MAD is 0.53 and 0.85,MSE is 0.54 and 1.15 respectively.By contrast,the precision of regularized RBF network model about predicted values is much higher.【Conclusion】 Selecting training sample and testing sample reasonably has provided a new
张少迪; 王延杰; 孙宏海
2012-01-01
A network training method for star pattern recognition was designed by combining a classific Radial Basic Function(RBF) neural network and star pattern samples. Firstly, the star pattern abstraction method was discussed and a triangulation based on star magnitudes was induced to connect the stars which probably appear in the same field of view. By taking extrated angular distances as the characteristic of star pattern, a star pattern sample set with completion, translation and rotation in-variance was established. Then, RBF neural network was studied to recognize the star patterns. RBF network training method was classified as sequence learning and batch learning. Some typical algorithms that could represent the two methods were studied on their advantages and disadvantages,and a new training method was designed based on the specialty of above star pattern sample sets. Experiments indicate that the designed method is more appropriate than those typical algorithms. Several star images were simulated through software, which was regarded as the observatory data and entered into the trained RBF neural network to test. The experiment results show that the network can recognize all the star patterns successfully.%根据经典径向基函数(RBF)神经网络的优势,结合星图模式样本集的特点,设计了一种适合星图模式样本的网络训练算法.从提取星图模式入手,引入三角剖分理论,将可能出现在同一视场内的恒星以三角形的形式连接起来,提取连接的角距作为星图模式,建立了具有完备性、平移旋转不变性的星图模式样本集.然后,利用RBF神经网络做星图识别,研究顺序训练方法和批量训练方法,总结多种经典算法的优缺点,并设计了一种训练方法.通过实验证明了该种方法较其他经典算法更为适合学习星图模式样本.最后,给出RBF神经网络相关的训练数据,并通过模拟星图软件获得若干模拟星图作为观测样本,利用
What influences children's conceptualizations of language input?
Plante, Elena; Vance, Rebecca; Moody, Amanda; Gerken, LouAnn
2013-10-01
Children learning language conceptualize the nature of input they receive in ways that allow them to understand and construct utterances they have never heard before. This study was designed to illuminate the types of information children with and without specific language impairment (SLI) focus on to develop their conceptualizations and whether they can rapidly shift their initial conceptualizations if provided with additional input. In 2 studies, preschool children with and without SLI were exposed to an artificial language, the characteristics of which allowed for various types of conceptualizations about its fundamental properties. After being familiarized with the language, children were asked to judge test strings that conformed to the input in 1 of 4 different ways. All children preferred test items that reflected a narrow conceptualization of the input (i.e., items most like those heard during familiarization). Children showed a strong preference for phonology as a defining property of the artificial language. Restructuring the input to the child could induce them to track word order information as well. Children tend toward narrow conceptualizations of language input, but the nature of their conceptualizations can be influenced by the nature of the input they receive.
陈红构; 赵晔
2011-01-01
A Multi-objective decision-making model is put forward to solve the issue of the supply chain cooperation partner selection in logistics enterprises. The paper excogitates a system of evaluation which is coincident with the characteristics of the logistics supply chain based on the essential feature and performance of logistics enterprises. Three aspects of logistics enterprise supply chain partners' partnership, partner characteristics and flexible cooperation are analyzed. The partners are classified as the core partners, important partners, potential partners, foundation partners. It is effectiveness and availability of this arithmetic that structuring the cooperation partner selection model in logistics enterprises based on the RBF neural networks.%提出了物流供应链企业合作伙伴选择问题的多目标决策模型,结合物流企业的行业特点,设计了符合物流企业供应链特点的合作伙伴选择指标体系,从伙伴关系、伙伴特性和合作柔性三个方面对物流企业供应链合作伙伴进行分析,将伙伴分类为核心伙伴、重要伙伴、潜力伙伴、基础伙伴,通过构造基于径向基函数神经网络的选择模型对合作伙伴进行归类选择.通过实证研究,结果表明该方法有效、实用.
U.S. Geological Survey, Department of the Interior — The data are input data files to run the forest simulation model Landis-II for Isle Royale National Park. Files include: a) Initial_Comm, which includes the location...
Existence conditions for unknown input functional observers
Fernando, T.; MacDougall, S.; Sreeram, V.; Trinh, H.
2013-01-01
This article presents necessary and sufficient conditions for the existence and design of an unknown input Functional observer. The existence of the observer can be verified by computing a nullspace of a known matrix and testing some matrix rank conditions. The existence of the observer does not require the satisfaction of the observer matching condition (i.e. Equation (16) in Hou and Muller 1992, 'Design of Observers for Linear Systems with Unknown Inputs', IEEE Transactions on Automatic Control, 37, 871-875), is not limited to estimating scalar functionals and allows for arbitrary pole placement. The proposed observer always exists when a state observer exists for the unknown input system, and furthermore, the proposed observer can exist even in some instances when an unknown input state observer does not exist.
Genetic search feature selection for affective modeling
Martínez, Héctor P.; Yannakakis, Georgios N.
2010-01-01
Automatic feature selection is a critical step towards the generation of successful computational models of affect. This paper presents a genetic search-based feature selection method which is developed as a global-search algorithm for improving the accuracy of the affective models built....... The method is tested and compared against sequential forward feature selection and random search in a dataset derived from a game survey experiment which contains bimodal input features (physiological and gameplay) and expressed pairwise preferences of affect. Results suggest that the proposed method...
Learning language in autism: maternal linguistic input contributes to later vocabulary.
Bang, Janet; Nadig, Aparna
2015-04-01
It is well established that children with typical development (TYP) exposed to more maternal linguistic input develop larger vocabularies. We know relatively little about the linguistic environment available to children with autism spectrum disorders (ASD), and whether input contributes to their later vocabulary. Children with ASD or TYP and their mothers from English and French-speaking families engaged in a 10 min free-play interaction. To compare input, children were matched on language ability, sex, and maternal education (ASD n = 20, TYP n = 20). Input was transcribed, and the number of word tokens and types, lexical diversity (D), mean length of utterances (MLU), and number of utterances were calculated. We then examined the relationship between input and children's spoken vocabulary 6 months later in a larger sample (ASD: n = 19, 50-85 months; TYP: n = 44, 25-58 months). No significant group differences were found on the five input features. A hierarchical multiple regression model demonstrated input MLU significantly and positively contributed to spoken vocabulary 6 months later in both groups, over and above initial language levels. No significant difference was found between groups in the slope between input MLU and later vocabulary. Our findings reveal children with ASD and TYP of similar language levels are exposed to similar maternal linguistic environments regarding number of word tokens and types, D, MLU, and number of utterances. Importantly, linguistic input accounted for later vocabulary growth in children with ASD.
Inadequacies of TPR and Krashen's Input Hypothesis
Meng Meng; LI Laifa
2008-01-01
In this paper,the rationale of TPR and the Input Hypothesis of Krashen which justifies practices of TPR are reviewed and criticized in the light of evidence from teachers'observation of a long-term TPR project.It is argued that the effectiveness of TPR is compromised by its inadequate attention to the complexity of classroom interactions and children's cognition.The Input Hypothesis is believed that it oversimplified the cognitive dynamics of language learning.
Land Scale, Input-Output and Income
Mengzhi; DENG
2013-01-01
Based on the investigation of production, inputs and income of tobacco farmers in 337 families in 10 counties of which the specialty is tobacco in Henan Province in 2010, the differences in the production, inputs and income were discussed. Results suggested that in terms of land yield rate and tobacco growers income, the suitable proportion of land for tobacco production in Henan Province is from 0.33 to 0.67 hm2.
Neural Networks with Complex and Quaternion Inputs
Rishiyur, Adityan
2006-01-01
This article investigates Kak neural networks, which can be instantaneously trained, for complex and quaternion inputs. The performance of the basic algorithm has been analyzed and shown how it provides a plausible model of human perception and understanding of images. The motivation for studying quaternion inputs is their use in representing spatial rotations that find applications in computer graphics, robotics, global navigation, computer vision and the spatial orientation of instruments. ...
Significance of input correlations in striatal function.
Man Yi Yim
2011-11-01
Full Text Available The striatum is the main input station of the basal ganglia and is strongly associated with motor and cognitive functions. Anatomical evidence suggests that individual striatal neurons are unlikely to share their inputs from the cortex. Using a biologically realistic large-scale network model of striatum and cortico-striatal projections, we provide a functional interpretation of the special anatomical structure of these projections. Specifically, we show that weak pairwise correlation within the pool of inputs to individual striatal neurons enhances the saliency of signal representation in the striatum. By contrast, correlations among the input pools of different striatal neurons render the signal representation less distinct from background activity. We suggest that for the network architecture of the striatum, there is a preferred cortico-striatal input configuration for optimal signal representation. It is further enhanced by the low-rate asynchronous background activity in striatum, supported by the balance between feedforward and feedback inhibitions in the striatal network. Thus, an appropriate combination of rates and correlations in the striatal input sets the stage for action selection presumably implemented in the basal ganglia.
李璞; 冯博
2016-01-01
With the rapid development of China's science and technology industry , the rapid development of science and technology industry , machinery automation , computer control system and the continuous development of the measurement industry , making the research of mobile robot has reached an unprecedented level , the robot has been widely used in ag-ricultural production ,industrial production ,National security ,life services and higher research design and other aspects of the design .As a part of the robot ,the mobile robot has concentrated the research results of intelligent sensing technology , mechanical manufacturing ,electronic communication technology ,intelligent instrument and automation control engineering . It is one of the most advanced technology research and design .In this paper , based on genetic algorithm to optimize the RBF network approximation algorithm , based on the characteristics of the robot motion trajectory , the robot motion trajec-tory control technology is studied .Through the real-time sensor online perception system , the robot can plan a trajectory without collision and route .The experimental results show that the design and research of the robot motion trajectory opti-mization technology , with good control effect , which can effectively avoid the obstacles in the course of the road , its reli-ability , good stability , the application prospect is very broad .%随着国力的不断增长，我国科技产业发展突飞猛进，机械自动化、计算机控制系统和测试计量行业的不断发展，使得移动机器人的研究也达到了一个前所未有的高度，机器人已经被广泛地应用到农业生产、工业生产、国家安全、生活服务和高等研究设计等领域的各个方面。移动机器人作为机器人的一部分，集中了智能传感技术、机械制造、电子无线通信技术、智能仪器和自动化控制工程等多学科的研究成果，是当前科技研究与设计最前沿
Adaptive sliding mode control of tethered satellite deployment with input limitation
Ma, Zhiqiang; Sun, Guanghui
2016-10-01
This paper proposes a novel adaptive sliding mode tension control method for the deployment of tethered satellite, where the input tension limitation is taken into account. The underactuated governing equations of the tethered satellites system are firstly derived based on Lagrangian mechanics theory. Considering the fact that the tether can only resist axial stretching, the tension input is modelled as input limitation. New adaptive sliding mode laws are addressed to guarantee the stability of the tethered satellite deployment with input disturbance, meanwhile to eliminate the effect of the limitation features of the tension input. Compared with the classic control strategy, the newly proposed adaptive sliding mode control law can deploy the satellite with smaller overshoot of the in-plane angle and implement the tension control reasonably and effectively in engineering practice. The numerical results validate the effectiveness of the proposed methods.
Determining the input dimension of a neural network for nonlinear time series prediction
张胜; 刘红星; 高敦堂; 都思丹
2003-01-01
Determining the input dimension of a feed-forward neural network for nonlinear time series prediction plays an important role in the modelling.The paper first summarizes the current methods for determining the input dimension of the neural network.Then inspired by the fact that the correlation dimension of a nonlinear dynamic system is the mostimportant feature of it,the paper presents a new idea that the input dimension of the neural network for nonlinear time series prediction can be taken as an integer just greater than or equal to the correlation dimension.Finally,some wlidation examples and results are given.
Input and output gain modulation by the lateral interhemispheric network in early visual cortex.
Wunderle, Thomas; Eriksson, David; Peiker, Christiane; Schmidt, Kerstin E
2015-05-20
Neurons in the cerebral cortex are constantly integrating different types of inputs. Dependent on their origin, these inputs can be modulatory in many ways and, for example, change the neuron's responsiveness, sensitivity, or selectivity. To investigate the modulatory role of lateral input from the same level of cortical hierarchy, we recorded in the primary visual cortex of cats while controlling synaptic input from the corresponding contralateral hemisphere by reversible deactivation. Most neurons showed a pronounced decrease in their response to a visual stimulus of different contrasts and orientations. This indicates that the lateral network acts via an unspecific gain-setting mechanism, scaling the output of a neuron. However, the interhemispheric input also changed the contrast sensitivity of many neurons, thereby acting on the input. Such a contrast gain mechanism has important implications because it extends the role of the lateral network from pure response amplification to the modulation of a specific feature. Interestingly, for many neurons, we found a mixture of input and output gain modulation. Based on these findings and the known physiology of callosal connections in the visual system, we developed a simple model of lateral interhemispheric interactions. We conclude that the lateral network can act directly on its target, leading to a sensitivity change of a specific feature, while at the same time it also can act indirectly, leading to an unspecific gain setting. The relative contribution of these direct and indirect network effects determines the outcome for a particular neuron.
Monitoring of Concrete Slab Dam Deformation Based on AFSA-RBF Neural Network%基于AFSA－RBF模型的混凝土平板坝变形监测
梁嘉琛; 赵鲲鹏; 杨景文
2016-01-01
针对混凝土平板坝水平位移监测序列呈非线性变化的特点，采用经验模态分解（ EMD）方法对混凝土平板坝水平位移监测序列进行分解，并采用计算最大信噪比的方法对信号进行去噪。面板坝的坝顶上下游方向水平位移主要受上下游水位和环境温度的影响，据此建立AFSA－RBF神经网络模型和RBF神经网络模型，对混凝土平板坝上下游方向水平位移进行预测，结果表明：AFSA－RBF模型能够很好地反映混凝土平板坝水平位移变化趋势和规律，预测结果有较高的精度，符合大坝安全监测的要求，可以在混凝土平板坝安全监测和评价中应用。%In view of the nonlinear and cyclical change characteristics of the horizontal displacement observation data of concrete slab dam, this paper used the method of empirical mode decomposition ( EMD) to decompose the horizontal displacement observation data of concrete slab dam and adopted the method of calculating the maximum SNR to denoise the signal. Concrete slab dam horizontal displacements of dam crest upstream and downstream direction mainly were affected by upstream and downstream water level and environment temperature. Based on this, it established AFSA⁃RBF and General RBF neural network model, and forecasted the horizontal displacement of concrete slab dam. The results show that AFSA⁃RBF model can reflect the tendency and rule of the concrete slab dam horizontal displacement change well. The predicted results have higher accuracy, which can meet the requirement of dam safety monitoring. The model can be applied in the safety monitoring and evaluation of concrete slab dam.
薛亮; 樊卫国; 汪小志
2016-01-01
In order to improve the accuracy of robot manipulator movement and improve the efficiency of robot movement, a methodis proposed based on genetic algorithm and RBF neural network .The robot manipulator's movement and the whole trajectory are optimized.In order to verify the design of the picking robot reliability, in the experimental green-house on the robot's picking performance were tested, test items include robot path planning of mobile and manipulator path planning.Through the test, we found that using the RBF neural network algorithm can effectively control of manipu-lator motion in the three-dimensional space, in under the control of the genetic algorithm, the robot can with less amount of calculation using neural network algorithm search to get the optimal path, and the calculation precision is above 99%, for its high accuracy, which provides a valuable reference for the fast computational efficiency and effect of high vegetable production picking robot design.%为了提高果蔬采摘机器人机械手运动的精确性，提高机器人移动的效率，提出了一种基于遗传算法和RBF网络的机器人运动轨迹控制方法，并对果蔬机器人机械手的活动和整体的移动轨迹进行优化，有效地提高了果蔬采摘机器人的工作精度和作业效率。为了验证设计的采摘机器人的可靠性，在大棚内对机器人的采摘性能进行了测试，包括机器人移动路径规划和机械手路径规划。通过测试发现：使用 RBF 神经网络算法可以有效地控制机械手在三维空间内的运动；在遗传算法控制下，机器人可以通过较少的计算次数利用神经网络算法搜索得到最优路径，计算精度达到了99％以上。其计算精度及效率高，为高效果蔬采摘机器人的设计提供了较有价值的参考。
GPS/leveling quasi-geoid fitting based on RBF neural networks%基于RBF神经网络的GPS/水准高程异常拟合
束蝉方; 李斐; 李明峰
2011-01-01
To determine an orthometric height using GPS, it is necessary to know the geoid/ quasi-geoid undulation. Fitting of GPS/leveling scatting data is one of main methods to get the quasi-geoid unknown. This paper proposed a new method for the fitting of GPS/leveling data based on radial basis functions (RBF) neural networks. The new learning algorithm selected the centers among the vertices of the Vornoni diagram of the sample data points. The bandwidth parameters of RBF, which was supposed experientially to be linear related to the distance from the center to the nearest scattering data point, were chosen optimizedly using generalized cross validation (GCV). The numerical results tested in one zone indicate that the new method is efficient for the geoid/quasi-geoid undulation fitting.%通过对离散GPS/水准点观测数据进行拟合从而获得区域内任意一点的高程异常是工程实践中经常遇到的问题.本文将径向基函数(RBF)神经网络方法应用于GPS/水准高程异常拟合,提出了一种新型网络学习方法.该方法首先通过对GPS/水准数据点进行Delaunay三角剖分,以其对偶Voronoi图的节点来构造选择基函数中心,再通过广义交叉验证(GCV)来最优确定基函数的宽度参数,最后利用最小二乘来确定RBF的输出权值,从而优化网络学习效果.实验结果表明,该学习方法取得良好的网络性能,和其它常用拟合方法的比较结果也反映出RBF神经网络适合应用于GPS/水准高程异常拟合.
王军号; 孟祥瑞; 吴宏伟
2011-01-01
针对瓦斯传感器常见的偏置型、冲击型、漂移型和周期型4种突发型故障,以小波分析和RBF神经网络为基础,提出了由小波包分解提取特征能量谱与扩展Kalman滤波算法(EKF)优化的RBF神经网络进行模式分类辨识的瓦斯传感器故障诊断方法.对瓦斯传感器的输出信号进行小波包分解,运用基于代价函数的局域判别基(LDB)算法进行裁剪,获取最优的特征能量谱,经处理后作为特征向量训练EKF-RBF神经网络,采用参数增广和统计动力学方法,通过带有整定因子的EKF参数估计,用来辨识瓦斯传感器的故障类型.实验结果表明:该方法的辨识正确率在95%以上,误报率和漏报率都明显优于其他算法,能够有效用于瓦斯传感器的故障在线诊断.%For four types of common abrupt faults of gas sensor, namely offset, impact, drift and periodic types, on the basis of wavelet analysis and RBF neural network, a method of the gas sensor fault diagnosis was proposed based on the pattern classification of characteristic energy spectrum extracted by the decomposition of wavelet packet and RBF neural network optimized by Extended Kalman Filter (EKF). The optimal characteristic energy spectrum was obtained through the decomposition of wavelet packet of output signal of gas sensor and optimally cut by Local Discriminant Base (LDB) based on the cost function. After processed, as the characteristic vector for training EKF-RBF neural network, adopted augnented parameters and method of statistical mechanics, and through the EKF parameter estimation with tuning factor, it was used to identify the fault type of sensor. The experimental results show that the identification accuracy is above 95 ％ ,its rate of false alarm and fail alarm is superior to other algorithms, and the method can be effectively applied to the online fault diagnosis of gas sensor.
基于RBF神经网络的池州市降水序列预测%Prediction of precipitation series based on RBF neural network in Chizhou city
沈艳; 杨春雷; 张庆国; 朱雅莉
2012-01-01
时间序列预测分析方法是进行预测预报的有效工具,有着广泛的应用.针对时间序列的非线性、动态变化等特征,基于RBF神经网络对时间序列预测方法进行改进,并以安徽省池州市1959 ～2009年来的月降水量为时间序列数据样本,用MATLAB软件编程,采用基于随机选取中心的RBF神经网络预测方法,对池州市的月降水量进行预测,并选择不同的扩展速度参数,用均方误差进行检验.通过与BP网络模型的预测结果比较分析,表明RBF模型的预测效果较好.建立的基于随机选取中心的RBF神经网络模型,不需要计算原始时间序列数据的复杂函数关系,具有操作简单、学习速度快、短期预测精度高等优点,用于时间序列预测方面能够获得十分满意的结果,具有很高的应用价值.%The time series analysis which used to forecast time series is an effective method and has been used widely. For the non-linear and dynamic characters, we improved time series forecast method based on RBF neural network with month precipitation of 1959-2009 in Chizhou, Anhui as data sample of time series. We used MATLAB software to program and forecast month precipitation of Chizhou based on RBF neural network forecast method of random selection center. At last, we chose different speed and test it with RMSE (root mean square error) . Compared with BPNN(backpropagation neural network), predictive validity of RBF model based on random selection center is preferable and does not need to calculate complexed functional relationship of the original time series data, with advantage of simple operation, study fast and high short-term forecast accuracy.
基于混沌径向基函数的风电功率短期预测%Prediction of short-term wind power based on chaotic RBF
李玲玲; 李宗礼; 李俊豪; 李志刚
2014-01-01
风电功率预测方法分为两类，即直接预测法与功率曲线转换法。因风电功率具有混沌特性，故将混沌时间序列的相关理论引入到风速和风电功率预测中。鉴于预测精度在很大程度上取决于模型参数的选择，为此先用C-C法联合优化了重构相空间的参数，再用径向基RBF神经网络模型直接预测风电功率，或者由该模型得到风速预测值后，根据对应的风电机组功率特性曲线而推算出风电功率预测值。实例分析结果表明：所提出的两种方法均有较高的预测精度，其中基于混沌径向基RBF神经网络的风电功率直接预测法效果更优。%There were two kinds of method to predict the wind power, which were the direct forecasting method and the power curve conversion forecasting method. The wind speed and the wind power were predicted with the theories related to the chaotic time series because the wind power was chaotic. First the parameters of the phase-space reconstruction were optimized by C-C method because the accuracy of the prediction largely depended on the parameters used; then the wind power was predicted by the RBF neural network. The predicted wind power could also be achieved based on the curve of wind turbines after the wind speed was predicted by the RBF neural network. The analysis of the example shows that both of the methods have good performances in accuracy and the direct way based on the chaotic RBF neural network is better than the other one.
Ant-cuckoo colony optimization for feature selection in digital mammogram.
Jona, J B; Nagaveni, N
2014-01-15
Digital mammogram is the only effective screening method to detect the breast cancer. Gray Level Co-occurrence Matrix (GLCM) textural features are extracted from the mammogram. All the features are not essential to detect the mammogram. Therefore identifying the relevant feature is the aim of this work. Feature selection improves the classification rate and accuracy of any classifier. In this study, a new hybrid metaheuristic named Ant-Cuckoo Colony Optimization a hybrid of Ant Colony Optimization (ACO) and Cuckoo Search (CS) is proposed for feature selection in Digital Mammogram. ACO is a good metaheuristic optimization technique but the drawback of this algorithm is that the ant will walk through the path where the pheromone density is high which makes the whole process slow hence CS is employed to carry out the local search of ACO. Support Vector Machine (SVM) classifier with Radial Basis Kernal Function (RBF) is done along with the ACO to classify the normal mammogram from the abnormal mammogram. Experiments are conducted in miniMIAS database. The performance of the new hybrid algorithm is compared with the ACO and PSO algorithm. The results show that the hybrid Ant-Cuckoo Colony Optimization algorithm is more accurate than the other techniques.
Ensemble classification of colon biopsy images based on information rich hybrid features.
Rathore, Saima; Hussain, Mutawarra; Aksam Iftikhar, Muhammad; Jalil, Abdul
2014-04-01
In recent years, classification of colon biopsy images has become an active research area. Traditionally, colon cancer is diagnosed using microscopic analysis. However, the process is subjective and leads to considerable inter/intra observer variation. Therefore, reliable computer-aided colon cancer detection techniques are in high demand. In this paper, we propose a colon biopsy image classification system, called CBIC, which benefits from discriminatory capabilities of information rich hybrid feature spaces, and performance enhancement based on ensemble classification methodology. Normal and malignant colon biopsy images differ with each other in terms of the color distribution of different biological constituents. The colors of different constituents are sharp in normal images, whereas the colors diffuse with each other in malignant images. In order to exploit this variation, two feature types, namely color components based statistical moments (CCSM) and Haralick features have been proposed, which are color components based variants of their traditional counterparts. Moreover, in normal colon biopsy images, epithelial cells possess sharp and well-defined edges. Histogram of oriented gradients (HOG) based features have been employed to exploit this information. Different combinations of hybrid features have been constructed from HOG, CCSM, and Haralick features. The minimum Redundancy Maximum Relevance (mRMR) feature selection method has been employed to select meaningful features from individual and hybrid feature sets. Finally, an ensemble classifier based on majority voting has been proposed, which classifies colon biopsy images using the selected features. Linear, RBF, and sigmoid SVM have been employed as base classifiers. The proposed system has been tested on 174 colon biopsy images, and improved performance (=98.85%) has been observed compared to previously reported studies. Additionally, the use of mRMR method has been justified by comparing the
Online feature selection with streaming features.
Wu, Xindong; Yu, Kui; Ding, Wei; Wang, Hao; Zhu, Xingquan
2013-05-01
We propose a new online feature selection framework for applications with streaming features where the knowledge of the full feature space is unknown in advance. We define streaming features as features that flow in one by one over time whereas the number of training examples remains fixed. This is in contrast with traditional online learning methods that only deal with sequentially added observations, with little attention being paid to streaming features. The critical challenges for Online Streaming Feature Selection (OSFS) include 1) the continuous growth of feature volumes over time, 2) a large feature space, possibly of unknown or infinite size, and 3) the unavailability of the entire feature set before learning starts. In the paper, we present a novel Online Streaming Feature Selection method to select strongly relevant and nonredundant features on the fly. An efficient Fast-OSFS algorithm is proposed to improve feature selection performance. The proposed algorithms are evaluated extensively on high-dimensional datasets and also with a real-world case study on impact crater detection. Experimental results demonstrate that the algorithms achieve better compactness and higher prediction accuracy than existing streaming feature selection algorithms.
DC SQUIDS with planar input coils
Pegrum, C.M.; Donaldson, G.B.; Hutson, D.; Tugwell, A.
1985-03-01
We describe the key parts of our recent work to develop a planar thin-film DC SQUID with a closely-coupled spiral input coil. Our aim has been to make a device that is superior to present RF SQUID sensors in terms of sensitivity and long-term reliability. To be compatible with an RF SQUID the inductance of the input coils must be relatively large, typically 2..mu..H, and the input noise current in the white noise region should be below 10pA Hz /SUP -1/2/ . A low level of 1/f noise is also necessary for many applications and should be achieved without the use of complex noisecancelling circuitry. Our devices meet these criteria. We include a description of work on window and edge junction fabrication using ion beam cleaning, thermal oxidation and RF plasma processing.
Harmonize input selection for sediment transport prediction
Afan, Haitham Abdulmohsin; Keshtegar, Behrooz; Mohtar, Wan Hanna Melini Wan; El-Shafie, Ahmed
2017-09-01
In this paper, three modeling approaches using a Neural Network (NN), Response Surface Method (RSM) and response surface method basis Global Harmony Search (GHS) are applied to predict the daily time series suspended sediment load. Generally, the input variables for forecasting the suspended sediment load are manually selected based on the maximum correlations of input variables in the modeling approaches based on NN and RSM. The RSM is improved to select the input variables by using the errors terms of training data based on the GHS, namely as response surface method and global harmony search (RSM-GHS) modeling method. The second-order polynomial function with cross terms is applied to calibrate the time series suspended sediment load with three, four and five input variables in the proposed RSM-GHS. The linear, square and cross corrections of twenty input variables of antecedent values of suspended sediment load and water discharge are investigated to achieve the best predictions of the RSM based on the GHS method. The performances of the NN, RSM and proposed RSM-GHS including both accuracy and simplicity are compared through several comparative predicted and error statistics. The results illustrated that the proposed RSM-GHS is as uncomplicated as the RSM but performed better, where fewer errors and better correlation was observed (R = 0.95, MAE = 18.09 (ton/day), RMSE = 25.16 (ton/day)) compared to the ANN (R = 0.91, MAE = 20.17 (ton/day), RMSE = 33.09 (ton/day)) and RSM (R = 0.91, MAE = 20.06 (ton/day), RMSE = 31.92 (ton/day)) for all types of input variables.
黄榕波; 郭穗勋
2013-01-01
建立基于径向基函数网络(Radial Basis Function network,RBF)的个体差异与药代动力学参数之间的关系模型.应用主成分分析(Principal Component Analysis,PCA)从个体差异因子中获取主成分作为RBF的输入,降低RBF的输入维数,从而降低了系统的复杂性.通过健康志愿者获得的实验数据对模型进行测试,实验结果表明RBF模型具有良好的拟合能力.
基于RBF神经网络的水下机器人传感器状态监测方法研究%Research on condition monitor for AUV sensors based on RBF neural network
张铭钧; 孙瑞琛; 王玉甲
2005-01-01
为了实现水下机器人多传感器状态监测,根据其工作环境及所配置传感器的数量,提出了基于径向基函数(RBF)神经网络的传感器状态监测方法,建立了二级神经网络监测模型,解决了多传感器故障诊断和信号恢复的问题.基于某型水下机器人海中试验数据进行计算机仿真试验的结果,验证了该方法的有效性和可行性.
陈建松; 闵雁; 伍乃骐
2008-01-01
本文在CAE仿真的基础上,采用田口试验设计(Taguchi)和径向基神经网络(RBF)技术对引起翘曲的塑参数进行了优化.结果表明:Tagucbi技术可以在较少的试验次数的情况下,确定各因素对翘曲的影响程度,获取各因素最佳的水平组合;运用RBF建立翘曲产生的神经网络模型,可以预测各因素在不同水平组合下的翘曲变形程度,达到离线预测的目的.
姬晓飞; 申东日; 陈义俊
2003-01-01
针对径向基函数(RBF)神经网络用于非线性系统辨识时存在的问题,对径向基函数网络的拓扑结构作了改进,并给出了改进的径向基函数(MRBF)神经网络的中心选取方法和权值在线调整算法,最后用改进的径向基函数网络对一个典型工业对象(CSTR)进行了应用研究,结果表明方法有效.
Forecast of US dollar/RMB exchange rate based on RBF neural network%基于RBF神经网络美元兑人民币汇率的预测
肖强; 钱晓东
2009-01-01
提出一种基于RBF神经网络的美元兑人民币汇率预测模型,该模型通过对近两年美元兑人民币汇率的历史数据分析,采用了改进的K-均值聚类算法,动态地确定RBF神经网络中心,并采用最小二乘法进行RBF神经网络的权值调整,通过美元兑人民币汇率的预测,结果表明该模型有较好的预测和泛化能力,可以取得好的预测结果.
黎云汉; 朱善安
2007-01-01
本文提出了一种基于递归正交最小二乘的径向基函数(RBF)网络人脸识别算法,该算法首先使用主成分分析(PCA)提取输入图像特征,将提取的特征作为RBF网络的输入进行识别,在求取网络权值时采用递归正交最小二乘(ROLS)算法.实验表明,该算法能明显地缩短训练时间同时具有较高的识别率.
RBF递归神经网络在供热解耦控制中的应用%Application of RBF recursion neural network to heat supply decoupling control
陈烈; 朱学莉; 齐维贵; 方修睦
2010-01-01
针对供热过程耦合特性和节能控制的需要,提出了一种基于径向基函数(RBF)递归神经网络的供热解耦控制方法.通过典型信号响应与最小二乘结合的方法得到供热耦合系统模型,利用RBF递归神经网络进行解耦控制,消除了质调节、量调节通道间的非线性强耦合作用.仿真结果证明该方法具有良好的解耦控制特性,满足供热系统多回路控制的要求.
雷金莉; 窦满峰
2014-01-01
Because of the environmental parameters transformation, the parameters perturbation and load torque disturbances of the brushless direct current motor ( BLDCM) in near space will appear, and the response speed and stability of control system will be bad. To solve this problem, we propose an adaptive fuzzy control algorithm based on RBF( radial basis function) neural network compensation. The adaptive fuzzy controller is deduced to ensure the BLDCM system has good dynamic performance, the RBF neural network is adopted to do online identification and compensate for the speed error when the parameters perturbation and load torque disturbance appear in order to a-chieve the purposes of fast response speed and good robustness. Comparing the simulation results of adaptive fuzzy control with those of RBF neural network compensation and adaptive fuzzy control, we show preliminarily that:(1) the adaptive fuzzy control Based on RBF neural network has a strong robustness against the uncertainties of the BLDCM;(2) its response time is shorten by adaptive fuzzy control over 10ms;(3) its peak electromagnetic torque is decreased about 20% during the response process.%近空间用无刷直流电机（ BLDCM）受环境参数影响出现不确定性参数摄动和负载扰动，系统的控制性能降低。为消除不确定性因素的影响，提出了一种基于RBF网络补偿的自适应模糊控制算法。该控制算法是在自适应模糊控制的基础上，引入RBF网络补偿控制器，对参数摄动和负载转矩突变引起的转速误差进行在线辨识和动态补偿，以达到快速鲁棒自适应控制目的。对比具有RBF网络补偿的自适应模糊控制和自适应模糊控制的模拟仿真实验结果表明：在转速变化、负载转矩突变和转动惯量改变条件下，有RBF网络补偿控制的响应时间缩短了10 ms以上，响应过程中，电磁转矩的瞬时峰值减少了20％左右，对近空间BLDCM系统的不确定性鲁棒性强。
Nonlinearities with Non-Gaussian Inputs.
1978-03-01
possessing a spectral density function . a constant. Then Jet arc tan [G(t)J be the input. By Theorem 3 this input is not bandlimited; and if The rando...such that the absolute ,,~~ovalue of any point in the spectrum is less than N. If the Gaussian process X(t) possesses a H ~ ) spectral density function (i.e...Gaussian process and th. series ii convergent pointvise as veil X(t ) possesses a spectral density function . as in an sense (51. Let z~( ) and g2
Input/Output Subroutine Library Program
Collier, James B.
1988-01-01
Efficient, easy-to-use program moved easily to different computers. Purpose of NAVIO, Input/Output Subroutine Library, provides input/output package of software for FORTRAN programs that is portable, efficient, and easy to use. Implemented as hierarchy of libraries. At bottom is very small library containing only non-portable routines called "I/O Kernel." Design makes NAVIO easy to move from one computer to another, by simply changing kernel. NAVIO appropriate for software system of almost any size wherein different programs communicate through files.
肖金凤; 肖杞铭; 曾铁军
2016-01-01
The available torque ripple suppression method of brushless DC motor has unsatisfactory suppression effect,or good ripple suppression effect while owning complicated learning algorithm,and is bad for promotion. To solve these problems, the field⁃oriented control system of the brushless DC motor was designed,which combines the RBF neural network and field⁃oriented control,and uses Luminary 615 microcontroller and brushless motor dedicated chip MC33035. The Visualbasic⁃based upper computer monitoring system matching with the motor was developed,which can realize the graphical display of parameters such as rotate speed and setting of motor parameters with low cost. The experimental results show that the designed RBF field⁃oriented control system of brushless DC motor has small torque ripple and high control accuracy.%针对现有无刷直流电机转矩脉动抑制方法存在抑制效果不理想，或脉动抑制效果好但学习算法复杂，不利于推广的问题，将RBF神经网络与磁场定向控制相结合，选用Luminary 615微控制器和无刷电机专用芯片MC33035，设计了无刷直流电机磁场定向控制系统。并开发基于Visual Basic的配套电机上位机监控系统，能在低成本下实现转速等参数的图形化显示及电机参数等的设置。实验结果表明，所设计的无刷直流电机RBF磁场定向控制系统转矩脉动小、控制精度高。
李建龙; 陈向东; 倪进权; 谢冰青
2014-01-01
RBF神经网络具有较强的拟合能力和稳定性，得到了广泛的应用。以FPGA芯片为核心器件，设计实现RBF神经网络。利用SOPC Builder设计硬件架构，通过添加指令，在NIOS环境下利用C语言进行设计，这样就解决了利用Verilog或VHDL设计消耗资源多和软件模拟耗时多的问题。最后以Altera公司的Cyclone IV系列芯片作为验证器件，结果表明该方法实现简单，可靠性强，消耗资源少。%RBF neural network with fitting ability and stability has been widely used. Based on the FPGA chip as a core de-vice,the RBF neural network is designed in this paper. The hardware architecture was designed by means of SOPC Builder,the added instructions and C language in NIOS environment. In this way,the problems existing in the design were solved,because they consume too many resources by using Verilog or VHDL to carry out the design and take much more time in software simula-tion. The Cyclone IV series chip of Altera Company was taken to perform the verification. The result shows that the method is simple,and has high reliability and less consumption of resources.
Prediction of the yield of biomass semi-coke based on RBF neural network%基于RBF神经网络生物质半焦产量的预测
胡爱娟; 刘杨; 袁清泉; 王桂荣; 石硕
2012-01-01
Biomass, as a renewable clean energy, has broad development prospects. Pyrolysis is the key step in thermochemical process of the biomass. This paper analyzes the factors which influence the semi-coke yield in the process of furfural residue and rice husk co-pyrolysis and sets up a RBF network model according to the three main factors to predict the yield of biomass semi-coke. The deviation of simulation results derive partly from the network and the learning samples and the selection of impact factors and more derive from experimental data. Through the error analysis, this model has better precision. The prediction results prove the feasibility of the application of RBF to thermo gravimetric pyrolysis of biomass.%生物质能作为一种可再生的清洁能源,开发利用前景广阔,热解反应是生物质热化学转化中的关键环节.本文分析了糠醛渣和稻壳共热解过程中影响半焦产量的各因素,根据三个主要因素,建立RBF神经网络模型,并应用此模型对半焦产量进行预测.其中,模拟结果的偏差一部分来源于网络和学习样本及影响因子的选取,更多的来源于实验数据.通过误差分析,本文建立的模型具有较好的精度,证明了RBF网络应用于热重分析仪中生物质热解领域的可行性.
庄述燕
2013-01-01
To install the reactive power compensation in grid connection point is an effective method to improve the voltage stability of wind power and ensure reliable operation. This paper designs a kind of STATCOM with superconducting magnetic energy storage. It achieves system decoupling of the feedback linearization using the inverse system method, and then controls the decoupled inverse system based on the RBF neural network sliding mode control. Simulation results show that the inverse system method and RBF neural network sliding mode control show a good effect in improving the system dynamic response speed and the system robustness. The results accord with the dynamic response characteristics of STATCOM with SMES.% 在并网点接入无功补偿器是提高电压稳定保证风电可靠运行的有效方法。设计一种带超导储能的静态无功补偿器，应用逆系统方法实现系统反馈线性化解耦，再采用RBF神经网络滑模控制对解耦逆系统进行控制。仿真结果表明：采用逆系统方法和RBF神经网络滑模控制对提高系统的动态响应速度和改善系统鲁棒性具有良好的效果，所得结果符合带超导储能的静态无功补偿器的动态响应特性。
基于PSO优化RBF神经网络的反应釜故障诊断%Application of PSO —based RBF Neural Network in Fault Diagnosis of CSTR
陈波; 潘海鹏; 邓志辉
2012-01-01
A new PSO algorithm with dynamically changing inertia weight and study factors based on improved adaptive PSO was proposed, where the inertia weight of the particle was adjusted adap-tively based on fitness of the particle. The diversity of inertia weight made a compromise between the global convergence and local convergence speed, so it can alleviate the problem of premature convergence effectively. The algorithm was applied to train RBF neural network and a model of fault diagnosis for CSTR was established,compared with PSO algorithm,the proposed algorithm can improve the training efficiency of neural network effectively and obtain good diagnosis results.%针对单一径向基函数(RBF)神经网络在反应釜故障诊断中泛化能力不足的缺点,设计了基于粒子群(PSO)算法优化的RBF神经网络.利用PSO算法操作简单、容易实现等特点及其智能背景,对RBF神经网络的参数、连接权重进行优化,并用经PSO算法优化的RBF神经网络对反应釜故障进行仿真诊断.仿真诊断结果表明,PSO算法优化的RBF神经网络具有较好的分类效果,较RBF诊断模型精度高、收敛快,具有推广应用价值.
An Analysis of Input Hypothesis in English Teaching
赖菲菲
2016-01-01
Input plays a significant role in the process of foreign language teaching and learning. One of the most important studies about input is Krashen's Input Hypothesis, which emphasizes the importance of comprehensive input in foreign language teaching and learning. This paper aims to study the significance of Input Hypothesis and its application to English teaching.
Declarative Semantics of Input Consuming Logic Programs
Bossi, Annalisa; Cocco, Nicoletta; Etalle, Sandro; Rossi, Sabina; Bruynooghe, Maurice; Lau, Kung-Kia
2004-01-01
Most logic programming languages actually provide some kind of dynamic scheduling to increase the expressive power and to control execution. Input consuming derivations have been introduced to describe dynamic scheduling while abstracting from the technical details. We review and compare the differe
Input and age effects: Quo vadis?
Weerman, F.
2014-01-01
The article discusses the important role played input in language acquisition. Topics include the difficulty in obtaining the difference between groups and languages, the visibility of the success of children concerning inflection in their knowledge, and the description of lexical mistake for monoli
The Contrast Theory of negative input.
Saxton, M
1997-02-01
Beliefs about whether or not children receive corrective input for grammatical errors depend crucially on how one defines the concept of correction. Arguably, previous conceptualizations do not provide a viable basis for empirical research (Gold, 1967; Brown & Hanlon, 1970; Hirsh-Pasek, Treiman & Schneiderman, 1984). Within the Contrast Theory of negative input, an alternative definition of negative evidence is offered, based on the idea that the unique discourse structure created in the juxtaposition of child error and adult correct form can reveal to the child the contrast, or conflict, between the two forms, and hence provide a basis for rejecting the erroneous form. A within-subjects experimental design was implemented for 36 children (mean age 5;0), in order to compare the immediate effects of negative evidence with those of positive input, on the acquisition of six novel irregular past tense forms. Children reproduced the correct irregular model more often, and persisted with fewer errors, following negative evidence rather than positive input.
Capital Power:From Input to Output
You Wanlong; Alice
2009-01-01
@@ After thirty yeas "going out" of China overseas investment,we learn from our failed lessons and also successful experience.Chinese enterprises are now standing at a new starting point of "going out".China is transforming from "capital input power" to "capital output power".
Treatments of Precipitation Inputs to Hydrologic Models
Hydrological models are used to assess many water resources problems from agricultural use and water quality to engineering issues. The success of these models are dependent on correct parameterization; the most sensitive being the rainfall input time series. These records can come from land-based ...
Input and Intake in Language Acquisition
Gagliardi, Ann C.
2012-01-01
This dissertation presents an approach for a productive way forward in the study of language acquisition, sealing the rift between claims of an innate linguistic hypothesis space and powerful domain general statistical inference. This approach breaks language acquisition into its component parts, distinguishing the input in the environment from…
Input and Intake in Language Acquisition
Gagliardi, Ann C.
2012-01-01
This dissertation presents an approach for a productive way forward in the study of language acquisition, sealing the rift between claims of an innate linguistic hypothesis space and powerful domain general statistical inference. This approach breaks language acquisition into its component parts, distinguishing the input in the environment from…
Declarative Semantics of Input Consuming Logic Programs
Bossi, Annalisa; Cocco, Nicoletta; Etalle, Sandro; Rossi, Sabina; Bruynooghe, Maurice; Lau, Kung-Kia
2004-01-01
Most logic programming languages actually provide some kind of dynamic scheduling to increase the expressive power and to control execution. Input consuming derivations have been introduced to describe dynamic scheduling while abstracting from the technical details. We review and compare the differe
Programmable Input for Nanomagnetic Logic Devices
Schmitt-Landsiedel D.
2013-01-01
Full Text Available A programmable magnetic input, based on the magnetic interaction of a soft and hard magnetic layer is presented for the ﬁrst time. Therefore, a single-domain Co/Pt nanomagnet is placed on top of one end of a permalloy bar, separated by a thin dielectric layer. The permalloy bar of the introduced input structure is magnetized by weak easy-axis in-plane ﬁelds. Acting like a ’magnetic ampliﬁer’, the generated fringing ﬁelds of the permalloy pole are strong enough to control the magnetization of the superimposed Co/Pt nanomagnets, which have high crystalline perpendicular magnetic anisotropy. This magnetostatic interaction results in a shift of the hysteresis curve of the Co/Pt nanomagnet, measured by magneto-optical Kerr microscopy. The Co/Pt nanomagnet is ﬁxed by the fringing ﬁeld of the permalloy and thereby not affected by the magnetic power clock of the Nanomagnetic Logic system. MFM measurements verify the functionality of the programmable magnetic input structure. The fringing ﬁelds are extracted from micromagnetic simulations and are in good agreement with experimental results. The introduced input structure enables switching the logic functionality of the majority gate from NAND to NOR during runtime, offering programmable Nanomagnetic Logic.
Sijbom, R.B.L.; Janssen, O.; van Yperen, N.W.
2015-01-01
We identified leaders’ achievement goals and composition of creative input as important factors that can clarify when and why leaders are receptive to, and supportive of, subordinates’ creative input. As hypothesized, in two experimental studies, we found that relative to mastery goal leaders, perfo
Sijbom, R.B.L.; Janssen, O.; van Yperen, N.W.
2015-01-01
We identified leaders’ achievement goals and composition of creative input as important factors that can clarify when and why leaders are receptive to, and supportive of, subordinates’ creative input. As hypothesized, in two experimental studies, we found that relative to mastery goal leaders, perfo
R.B.L. Sijbom; O. Janssen; N.W. van Yperen
2014-01-01
We identified leaders’ achievement goals and composition of creative input as important factors that can clarify when and why leaders are receptive to, and supportive of, subordinates’ creative input. As hypothesized, in two experimental studies, we found that relative to mastery goal leaders, perfo
Sijbom, R.B.L.; Janssen, O.; van Yperen, N.W.
2015-01-01
We identified leaders’ achievement goals and composition of creative input as important factors that can clarify when and why leaders are receptive to, and supportive of, subordinates’ creative input. As hypothesized, in two experimental studies, we found that relative to mastery goal leaders,
Ion-Mărgineanu, Adrian; Van Cauter, Sofie; Sima, Diana M.; Maes, Frederik; Sunaert, Stefan; Himmelreich, Uwe; Van Huffel, Sabine
2017-01-01
Purpose: The purpose of this paper is discriminating between tumor progression and response to treatment based on follow-up multi-parametric magnetic resonance imaging (MRI) data retrieved from glioblastoma multiforme (GBM) patients. Materials and Methods: Multi-parametric MRI data consisting of conventional MRI (cMRI) and advanced MRI [i.e., perfusion weighted MRI (PWI) and diffusion kurtosis MRI (DKI)] were acquired from 29 GBM patients treated with adjuvant therapy after surgery. We propose an automatic pipeline for processing advanced MRI data and extracting intensity-based histogram features and 3-D texture features using manually and semi-manually delineated regions of interest (ROIs). Classifiers are trained using a leave-one-patient-out cross validation scheme on complete MRI data. Balanced accuracy rate (BAR)–values are computed and compared between different ROIs, MR modalities, and classifiers, using non-parametric multiple comparison tests. Results: Maximum BAR–values using manual delineations are 0.956, 0.85, 0.879, and 0.932, for cMRI, PWI, DKI, and all three MRI modalities combined, respectively. Maximum BAR–values using semi-manual delineations are 0.932, 0.894, 0.885, and 0.947, for cMRI, PWI, DKI, and all three MR modalities combined, respectively. After statistical testing using Kruskal-Wallis and post-hoc Dunn-Šidák analysis we conclude that training a RUSBoost classifier on features extracted using semi-manual delineations on cMRI or on all MRI modalities combined performs best. Conclusions: We present two main conclusions: (1) using T1 post-contrast (T1pc) features extracted from manual total delineations, AdaBoost achieves the highest BAR–value, 0.956; (2) using T1pc-average, T1pc-90th percentile, and Cerebral Blood Volume (CBV) 90th percentile extracted from semi-manually delineated contrast enhancing ROIs, SVM-rbf, and RUSBoost achieve BAR–values of 0.947 and 0.932, respectively. Our findings show that AdaBoost, SVM-rbf, and
Fock, Eric
2014-08-01
A new algorithm for the selection of input variables of neural network is proposed. This new method, applied after the training stage, ranks the inputs according to their importance in the variance of the model output. The use of a global sensitivity analysis technique, extended Fourier amplitude sensitivity test, gives the total sensitivity index for each variable, which allows for the ranking and the removal of the less relevant inputs. Applied to some benchmarking problems in the field of features selection, the proposed approach shows good agreement in keeping the relevant variables. This new method is a useful tool for removing superfluous inputs and for system identification.
Recurrent network models for perfect temporal integration of fluctuating correlated inputs.
Hiroshi Okamoto
2009-06-01
Full Text Available Temporal integration of input is essential to the accumulation of information in various cognitive and behavioral processes, and gradually increasing neuronal activity, typically occurring within a range of seconds, is considered to reflect such computation by the brain. Some psychological evidence suggests that temporal integration by the brain is nearly perfect, that is, the integration is non-leaky, and the output of a neural integrator is accurately proportional to the strength of input. Neural mechanisms of perfect temporal integration, however, remain largely unknown. Here, we propose a recurrent network model of cortical neurons that perfectly integrates partially correlated, irregular input spike trains. We demonstrate that the rate of this temporal integration changes proportionately to the probability of spike coincidences in synaptic inputs. We analytically prove that this highly accurate integration of synaptic inputs emerges from integration of the variance of the fluctuating synaptic inputs, when their mean component is kept constant. Highly irregular neuronal firing and spike coincidences are the major features of cortical activity, but they have been separately addressed so far. Our results suggest that the efficient protocol of information integration by cortical networks essentially requires both features and hence is heterotic.
Minimum feature size preserving decompositions
Aloupis, Greg; Demaine, Martin L; Dujmovic, Vida; Iacono, John
2009-01-01
The minimum feature size of a crossing-free straight line drawing is the minimum distance between a vertex and a non-incident edge. This quantity measures the resolution needed to display a figure or the tool size needed to mill the figure. The spread is the ratio of the diameter to the minimum feature size. While many algorithms (particularly in meshing) depend on the spread of the input, none explicitly consider finding a mesh whose spread is similar to the input. When a polygon is partitioned into smaller regions, such as triangles or quadrangles, the degradation is the ratio of original to final spread (the final spread is always greater). Here we present an algorithm to quadrangulate a simple n-gon, while achieving constant degradation. Note that although all faces have a quadrangular shape, the number of edges bounding each face may be larger. This method uses Theta(n) Steiner points and produces Theta(n) quadrangles. In fact to obtain constant degradation, Omega(n) Steiner points are required by any al...
Hardwood species classification with DWT based hybrid texture feature extraction techniques
Arvind R Yadav; R S Anand; M L Dewal; Sangeeta Gupta
2015-12-01
In this work, discrete wavelet transform (DWT) based hybrid texture feature extraction techniques have been used to categorize the microscopic images of hardwood species into 75 different classes. Initially, the DWT has been employed to decompose the image up to 7 levels using Daubechies (db3) wavelet as decomposition filter. Further, first-order statistics (FOS) and four variants of local binary pattern (LBP) descriptors are used to acquire distinct features of these images at various levels. The linear support vector machine (SVM), radial basis function (RBF) kernel SVM and random forest classifiers have been employed for classification. The classification accuracy obtained with state-of-the-art and DWT based hybrid texture features using various classifiers are compared. The DWT based FOS-uniform local binary pattern (DWTFOSLBPu2) texture features at the 4th level of image decomposition have produced best classification accuracy of 97.67 ± 0.79% and 98.40 ± 064% for grayscale and RGB images, respectively, using linear SVM classifier. Reduction in feature dataset by minimal redundancy maximal relevance (mRMR) feature selection method is achieved and the best classification accuracy of 99.00 ± 0.79% and 99.20 ± 0.42% have been obtained for DWT based FOS-LBP histogram Fourier features (DWTFOSLBP-HF) technique at the 5th and 6th levels of image decomposition for grayscale and RGB images, respectively, using linear SVM classifier. The DWTFOSLBP-HF features selected with mRMR method has also established superiority amongst the DWT based hybrid texture feature extraction techniques for randomly divided database into different proportions of training and test datasets.
2011-09-12
.../ jurisdiction communications and lags far behind the commercial sector in data capability. Congressional...-device communication): A type of networking where each node must not only capture and disseminate its own... data in the network.\\5\\ \\5\\ http://en.wikipedia.org/wiki/Mesh_networking . Adaptability: The ability of...
J. Pavlovicova
2007-04-01
Full Text Available In this contribution, human face as biometric is considered. Original method of feature extraction from image data is introduced using MLP (multilayer perceptron and PCA (principal component analysis. This method is used in human face recognition system and results are compared to face recognition system using PCA directly, to a system with direct classification of input images by MLP and RBF (radial basis function networks, and to a system using MLP as a feature extractor and MLP and RBF networks in the role of classifier. Also a two-stage method for face recognition is presented, in which Kohonen self-organizing map is used as a feature extractor. MLP and RBF network are used as classifiers. In order to obtain deeper insight into presented methods, also visualizations of internal representation of input data obtained by neural networks are presented.
Protein sequence classification using feature hashing.
Caragea, Cornelia; Silvescu, Adrian; Mitra, Prasenjit
2012-06-21
Recent advances in next-generation sequencing technologies have resulted in an exponential increase in the rate at which protein sequence data are being acquired. The k-gram feature representation, commonly used for protein sequence classification, usually results in prohibitively high dimensional input spaces, for large values of k. Applying data mining algorithms to these input spaces may be intractable due to the large number of dimensions. Hence, using dimensionality reduction techniques can be crucial for the performance and the complexity of the learning algorithms. In this paper, we study the applicability of feature hashing to protein sequence classification, where the original high-dimensional space is "reduced" by hashing the features into a low-dimensional space, using a hash function, i.e., by mapping features into hash keys, where multiple features can be mapped (at random) to the same hash key, and "aggregating" their counts. We compare feature hashing with the "bag of k-grams" approach. Our results show that feature hashing is an effective approach to reducing dimensionality on protein sequence classification tasks.
Multiple actor-critic structures for continuous-time optimal control using input-output data.
Song, Ruizhuo; Lewis, Frank; Wei, Qinglai; Zhang, Hua-Guang; Jiang, Zhong-Ping; Levine, Dan
2015-04-01
In industrial process control, there may be multiple performance objectives, depending on salient features of the input-output data. Aiming at this situation, this paper proposes multiple actor-critic structures to obtain the optimal control via input-output data for unknown nonlinear systems. The shunting inhibitory artificial neural network (SIANN) is used to classify the input-output data into one of several categories. Different performance measure functions may be defined for disparate categories. The approximate dynamic programming algorithm, which contains model module, critic network, and action network, is used to establish the optimal control in each category. A recurrent neural network (RNN) model is used to reconstruct the unknown system dynamics using input-output data. NNs are used to approximate the critic and action networks, respectively. It is proven that the model error and the closed unknown system are uniformly ultimately bounded. Simulation results demonstrate the performance of the proposed optimal control scheme for the unknown nonlinear system.
Aarts, Rian; Demir-Vegter, Serpil; Kurvers, Jeanne; Henrichs, Lotte
2016-01-01
The current study examined academic language (AL) input of mothers and teachers to 15 monolingual Dutch and 15 bilingual Turkish-Dutch 4- to 6-year-old children and its relationships with the children's language development. At two times, shared book reading was videotaped and analyzed for academic features: lexical diversity, syntactic…
Indirect techniques for adaptive input-output linearization of non-linear systems
Teel, Andrew; Kadiyala, Raja; Kokotovic, Peter; Sastry, Shankar
1991-01-01
A technique of indirect adaptive control based on certainty equivalence for input output linearization of nonlinear systems is proven convergent. It does not suffer from the overparameterization drawbacks of the direct adaptive control techniques on the same plant. This paper also contains a semiindirect adaptive controller which has several attractive features of both the direct and indirect schemes.
Lacirignola, Martino; Blanc, Philippe; Girard, Robin; Pérez-López, Paula; Blanc, Isabelle
2017-02-01
In the life cycle assessment (LCA) context, global sensitivity analysis (GSA) has been identified by several authors as a relevant practice to enhance the understanding of the model's structure and ensure reliability and credibility of the LCA results. GSA allows establishing a ranking among the input parameters, according to their influence on the variability of the output. Such feature is of high interest in particular when aiming at defining parameterized LCA models. When performing a GSA, the description of the variability of each input parameter may affect the results. This aspect is critical when studying new products or emerging technologies, where data regarding the model inputs are very uncertain and may cause misleading GSA outcomes, such as inappropriate input rankings. A systematic assessment of this sensitivity issue is now proposed. We develop a methodology to analyze the sensitivity of the GSA results (i.e. the stability of the ranking of the inputs) with respect to the description of such inputs of the model (i.e. the definition of their inherent variability). With this research, we aim at enriching the debate on the application of GSA to LCAs affected by high uncertainties. We illustrate its application with a case study, aiming at the elaboration of a simple model expressing the life cycle greenhouse gas emissions of enhanced geothermal systems (EGS) as a function of few key parameters. Our methodology allows identifying the key inputs of the LCA model, taking into account the uncertainty related to their description.
Design and Implementation of Kana-Input Navigation System for Kids based on the Cyber Assistant
Hiroshi Matsuda
2004-02-01
Full Text Available In Japan, it has increased the opportunity for young children to experience the personal computer in elementary schools. However, in order to use computer, many domestic barriers have confronted young children (Kids because they cannot read difficult Kanji characters and had not learnt Roman alphabet yet. As a result, they cannot input text strings by JIS Kana keyboard. In this research, we developed Kana-Input NaVigation System for kids (KINVS based on the Cyber Assistant System (CAS. CAS is a Human-Style Software Robot based on the 3D-CG real-time animation and voice synthesis technology. KINVS enables to input Hiragana/Katakana characters by mouse operation only (without keyboard operation and CAS supports them by using speaking, facial expression, body action and sound effects. KINVS displays the 3D-Stage like a classroom. In this room, Blackboard, Interactive parts to input Kana-characters, and CAS are placed. As some results of preliminary experiments, it is definitely unfit for Kids to double-click objects quickly or to move the Scrollbar by mouse dragging. So, mouse input method of KINVS are designed to use only single click and wheeler rotation. To input characters, Kids clicks or rotates the Interactive Parts. KINVS reports all information by voice speaking and Kana subtitles instead of Kanji text. Furthermore, to verify the functional feature of KINVS, we measured how long Kids had taken to input long text by using KINVS.
Do efficiency scores depend on input mix?
Asmild, Mette; Hougaard, Jens Leth; Kronborg, Dorte
2013-01-01
In this paper we examine the possibility of using the standard Kruskal-Wallis (KW) rank test in order to evaluate whether the distribution of efficiency scores resulting from Data Envelopment Analysis (DEA) is independent of the input (or output) mix of the observations. Since the DEA frontier...... is estimated, many standard assumptions for evaluating the KW test statistic are violated. Therefore, we propose to explore its statistical properties by the use of simulation studies. The simulations are performed conditional on the observed input mixes. The method, unlike existing approaches...... the assumption of mix independence is rejected the implication is that it, for example, is impossible to determine whether machine intensive project are more or less efficient than labor intensive projects....
Multimodal interfaces with voice and gesture input
Milota, A.D.; Blattner, M.M.
1995-07-20
The modalities of speech and gesture have different strengths and weaknesses, but combined they create synergy where each modality corrects the weaknesses of the other. We believe that a multimodal system such a one interwining speech and gesture must start from a different foundation than ones which are based solely on pen input. In order to provide a basis for the design of a speech and gesture system, we have examined the research in other disciplines such as anthropology and linguistics. The result of this investigation was a taxonomy that gave us material for the incorporation of gestures whose meanings are largely transparent to the users. This study describes the taxonomy and gives examples of applications to pen input systems.
Controlling Synfire Chain by Inhibitory Synaptic Input
Shinozaki, Takashi; Câteau, Hideyuki; Urakubo, Hidetoshi; Okada, Masato
2007-04-01
The propagation of highly synchronous firings across neuronal networks, called the synfire chain, has been actively studied both theoretically and experimentally. The temporal accuracy and remarkable stability of the propagation have been repeatedly examined in previous studies. However, for such a mode of signal transduction to play a major role in processing information in the brain, the propagation should also be controlled dynamically and flexibly. Here, we show that inhibitory but not excitatory input can bidirectionally modulate the propagation, i.e., enhance or suppress the synchronous firings depending on the timing of the input. Our simulations based on the Hodgkin-Huxley neuron model demonstrate this bidirectional modulation and suggest that it should be achieved with any biologically inspired modeling. Our finding may help describe a concrete scenario of how multiple synfire chains lying in a neuronal network are appropriately controlled to perform significant information processing.
Virtual input device with diffractive optical element
Wu, Ching Chin; Chu, Chang Sheng
2005-02-01
As a portable device, such as PDA and cell phone, a small size build in virtual input device is more convenient for complex input demand. A few years ago, a creative idea called 'virtual keyboard' is announced, but up to now there's still no mass production method for this idea. In this paper we'll show the whole procedure of making a virtual keyboard. First of all is the HOE (Holographic Optical Element) design of keyboard image which yields a fan angle about 30 degrees, and then use the electron forming method to copy this pattern in high precision. And finally we can product this element by inject molding. With an adaptive lens design we can get a well correct keyboard image in distortion and a wilder fan angle about 70 degrees. With a batter alignment of HOE pattern lithography, we"re sure to get higher diffraction efficiency.
Neuroprosthetics and the science of patient input.
Benz, Heather L; Civillico, Eugene F
2017-01-01
Safe and effective neuroprosthetic systems are of great interest to both DARPA and CDRH, due to their innovative nature and their potential to aid severely disabled populations. By expanding what is possible in human-device interaction, these devices introduce new potential benefits and risks. Therefore patient input, which is increasingly important in weighing benefits and risks, is particularly relevant for this class of devices. FDA has been a significant contributor to an ongoing stakeholder conversation about the inclusion of the patient voice, working collaboratively to create a new framework for a patient-centered approach to medical device development. This framework is evolving through open dialogue with researcher and patient communities, investment in the science of patient input, and policymaking that is responsive to patient-centered data throughout the total product life cycle. In this commentary, we will discuss recent developments in patient-centered benefit-risk assessment and their relevance to the development of neural prosthetic systems.
Model based optimization of EMC input filters
Raggl, K; Kolar, J. W. [Swiss Federal Institute of Technology, Power Electronic Systems Laboratory, Zuerich (Switzerland); Nussbaumer, T. [Levitronix GmbH, Zuerich (Switzerland)
2008-07-01
Input filters of power converters for compliance with regulatory electromagnetic compatibility (EMC) standards are often over-dimensioned in practice due to a non-optimal selection of number of filter stages and/or the lack of solid volumetric models of the inductor cores. This paper presents a systematic filter design approach based on a specific filter attenuation requirement and volumetric component parameters. It is shown that a minimal volume can be found for a certain optimal number of filter stages for both the differential mode (DM) and common mode (CM) filter. The considerations are carried out exemplarily for an EMC input filter of a single phase power converter for the power levels of 100 W, 300 W, and 500 W. (author)
Solar wind-magnetosphere energy input functions
Bargatze, L.F.; McPherron, R.L.; Baker, D.N.
1985-01-01
A new formula for the solar wind-magnetosphere energy input parameter, P/sub i/, is sought by applying the constraints imposed by dimensional analysis. Applying these constraints yields a general equation for P/sub i/ which is equal to rho V/sup 3/l/sub CF//sup 2/F(M/sub A/,theta) where, rho V/sup 3/ is the solar wind kinetic energy density and l/sub CF//sup 2/ is the scale size of the magnetosphere's effective energy ''collection'' region. The function F which depends on M/sub A/, the Alfven Mach number, and on theta, the interplanetary magnetic field clock angle is included in the general equation for P/sub i/ in order to model the magnetohydrodynamic processes which are responsible for solar wind-magnetosphere energy transfer. By assuming the form of the function F, it is possible to further constrain the formula for P/sub i/. This is accomplished by using solar wind data, geomagnetic activity indices, and simple statistical methods. It is found that P/sub i/ is proportional to (rho V/sup 2/)/sup 1/6/VBG(theta) where, rho V/sup 2/ is the solar wind dynamic pressure and VBG(theta) is a rectified version of the solar wind motional electric field. Furthermore, it is found that G(theta), the gating function which modulates the energy input to the magnetosphere, is well represented by a ''leaky'' rectifier function such as sin/sup 4/(theta/2). This function allows for enhanced energy input when the interplanetary magnetic field is oriented southward. This function also allows for some energy input when the interplanetary magnetic field is oriented northward. 9 refs., 4 figs.
Sensory Synergy as Environmental Input Integration
Fady eAlnajjar
2015-01-01
Full Text Available The development of a method to feed proper environmental inputs back to the central nervous system (CNS remains one of the challenges in achieving natural movement when part of the body is replaced with an artificial device. Muscle synergies are widely accepted as a biologically plausible interpretation of the neural dynamics between the CNS and the muscular system. Yet the sensorineural dynamics of environmental feedback to the CNS has not been investigated in detail. In this study, we address this issue by exploring the concept of sensory synergy. In contrast to muscle synergy, we hypothesize that sensory synergy plays an essential role in integrating the overall environmental inputs to provide low-dimensional information to the CNS. We assume that sensor synergy and muscle synergy communicate using these low-dimensional signals. To examine our hypothesis, we conducted posture control experiments involving lateral disturbance with 9 healthy participants. Proprioceptive information represented by the changes on muscle lengths were estimated by using the musculoskeletal model analysis software SIMM. Changes on muscles lengths were then used to compute sensory synergies. The experimental results indicate that the environmental inputs were translated into the two dimensional signals and used to move the upper limb to the desired position immediately after the lateral disturbance. Participants who showed high skill in posture control were found to be likely to have a strong correlation between sensory and muscle signaling as well as high coordination between the utilized sensory synergies. These results suggest the importance of integrating environmental inputs into suitable low-dimensional signals before providing them to the CNS. This mechanism should be essential when designing the prosthesis’ sensory system to make the controller simpler
EMOWARS: INTERACTIVE GAME INPUT MENGGUNAKAN EKSPRESI WAJAH
Andry Chowanda
2013-11-01
opportunity for researchers in affective game with a more interactive game play as well as rich and complex story. Hopefully this will improve the user affective state and emotions in game. The results of this research imply that happy emotion obtains 78% of detection, meanwhile the anger emotion has the lowest detection of 44.4%. Moreover, users prefer mouse and FER (face expression recognition as the best input for this game.
Cometary micrometeorites and input of prebiotic compounds
2014-01-01
The apparition of life on the early Earth was probably favored by inputs of extraterrestrial matter brought by carbonaceous chondrite-like objects or cometary material. Interplanetary dust collected nowadays on Earth is related to carbonaceous chondrites and to cometary material. They contain in particular at least a few percent of organic matter, organic compounds (amino-acids, PAHs,…), hydrous silicates, and could have largely contributed to the budget of prebiotic matter on Earth, about 4 ...
Emowars: Interactive Game Input Menggunakan Ekspresi Wajah
Andry Chowanda
2013-12-01
Full Text Available Research in the affective game has received attention from the research communities over this lustrum. As a crucial aspect of a game, emotions play an important role in user experience as well as to emphasize the user’s emotions state on game design. This will improve the user’s interactivity while they playing the game. This research aims to discuss and analyze whether emotions can replace traditional user game inputs (keyboard, mouse, and others. The methodology used in this research is divided into two main phases: game design and facial expression recognition. The results of this research indicate that users preferred to use a traditional input such as mouse. Moreover, user’s interactivities with game are still slightly low. However, this is a great opportunity for researchers in affective game with a more interactive game play as well as rich and complex story. Hopefully this will improve the user affective state and emotions in game. The results of this research imply that happy emotion obtains 78% of detection, meanwhile the anger emotion has the lowest detection of44.4%. Moreover, users prefer mouse and FER (face expression recognition as the best input for this game.
Molecular structure input on the web
Ertl Peter
2010-02-01
Full Text Available Abstract A molecule editor, that is program for input and editing of molecules, is an indispensable part of every cheminformatics or molecular processing system. This review focuses on a special type of molecule editors, namely those that are used for molecule structure input on the web. Scientific computing is now moving more and more in the direction of web services and cloud computing, with servers scattered all around the Internet. Thus a web browser has become the universal scientific user interface, and a tool to edit molecules directly within the web browser is essential. The review covers a history of web-based structure input, starting with simple text entry boxes and early molecule editors based on clickable maps, before moving to the current situation dominated by Java applets. One typical example - the popular JME Molecule Editor - will be described in more detail. Modern Ajax server-side molecule editors are also presented. And finally, the possible future direction of web-based molecule editing, based on technologies like JavaScript and Flash, is discussed.
Large-scale stabilization control of input-constrained quadrotor
Jun Jiang
2016-10-01
Full Text Available The quadrotor has been the most popular aircraft in the last decade due to its excellent dynamics and continues to attract ever-increasing research interest. Delivering a quadrotor from a large fixed-wing aircraft is a promising application of quadrotors. In such an application, the quadrotor needs to switch from a highly unstable status, featured as large initial states, to a safe and stable flight status. This is the so-called large-scale stability control problem. In such an extreme scenario, the quadrotor is at risk of actuator saturation. This can cause the controller to update incorrectly and lead the quadrotor to spiral and crash. In this article, to safely control the quadrotor in such scenarios, the control input constraint is analyzed. The key states of a quadrotor dynamic model are selected, and a two-dimensional dynamic model is extracted based on a symmetrical body configuration. A generalized point-wise min-norm nonlinear control method is proposed based on the Lyapunov function, and large-scale stability control is hence achieved. An enhanced point-wise, min-norm control is further provided to improve the attitude control performance, with altitude performance degenerating slightly. Simulation results showed that the proposed control methods can stabilize the input-constrained quadrotor and the enhanced method can improve the performance of the quadrotor in critical states.
Sparse polynomial surrogates for aerodynamic computations with random inputs
Savin, Eric; Peter, Jacques
2015-01-01
This paper deals with some of the methodologies used to construct polynomial surrogate models based on generalized polynomial chaos (gPC) expansions for applications to uncertainty quantification (UQ) in aerodynamic computations. A core ingredient in gPC expansions is the choice of a dedicated sampling strategy, so as to define the most significant scenarios to be considered for the construction of such metamodels. A desirable feature of the proposed rules shall be their ability to handle several random inputs simultaneously. Methods to identify the relative "importance" of those variables or uncertain data shall be ideally considered as well. The present work is more particularly dedicated to the development of sampling strategies based on sparsity principles. Sparse multi-dimensional cubature rules based on general one-dimensional Gauss-Jacobi-type quadratures are first addressed. These sets are non nested, but they are well adapted to the probability density functions with compact support for the random in...
Wernersson, Rasmus; Rapacki, Krzysztof; Stærfeldt, Hans Henrik
2006-01-01
FeatureMap3D is a web-based tool that maps protein features onto 3D structures. The user provides sequences annotated with any feature of interest, such as post-translational modifications, protease cleavage sites or exonic structure and FeatureMap3D will then search the Protein Data Bank (PDB...... without sequence annotation, to evaluate the quality of the alignment of the input sequences to the most homologous structures in the PDB, through the sequence conservation colored 3D structure visualization tool. FeatureMap3D is available at: http://www.cbs.dtu.dk/services/FeatureMap3D/....
The Importance of Input and Interaction in SLA
党春花
2009-01-01
As is known to us, input and interaction play the crucial roles in second language acquisition (SLA). Different linguistic schools have different explanations to input and interaction Behaviorist theories hold a view that input is composed of stimuli and response, putting more emphasis on the importance of input, while mentalist theories find input is a necessary condition to SLA, not a sufficient condition. At present, social interaction theories, which is one type of cognitive linguistics, suggests that besides input, interaction is also essential to language acquisition. Then, this essay will discuss how input and interaction result in SLA.
基于RB F神经网络的集成增量学习算法%RESEARCH ON RBF NEURAL NETWORK-BASED ENSEMBLE INCREMENTAL LEARNING ALGORITHM
彭玉青; 赵翠翠; 高晴晴
2016-01-01
针对增量学习的遗忘性问题和集成增量学习的网络增长过快问题，提出基于径向基神经网络（RBF）的集成增量学习方法。为了避免网络的遗忘性，每次学习新类别知识时都训练一个RBF神经网络，把新训练的RBF神经网络加入到集成系统中，从而组建成一个大的神经网络系统。分别采用最近中心法、最大概率法、最近中心与最大概率相结合的方法进行确定获胜子网络，最终结果由获胜子网络进行输出。在最大概率法中引入自组织映（SOM）的原型向量来解决类中心相近问题。为了验证网络的增量学习，用UCI机器学习库中Statlog（Landsat Satellite）数据集做实验，结果显示该网络在学习新类别知识后，既获得了新类别的知识也没有遗忘已学知识。%Aiming at the forgetfulness problem of incremental learning and the excessive network growth problem of the integrated incremental learning,this paper proposes an integrated incremental learning method which is based on the radial basis function (RBF)neural network.In order to avoid the forgetfulness of the network,for each knowledge learning of new category we all trained an RBF neural network,and added the newly trained RBF neural network to the integrated system so as to form a large system of neural networks.To determine the winning sub-network,we adopted the nearest centre method,the maximum probability method and the combination of these two methods,and the final result was outputted by the winning sub-network.Moreover,we introduced the prototype vectors of self-organising map to maximum probability method for solving the problem of class centre similarity.For verifying the proposed network incremental learning,we made the experiments using the Statlog (Landsat Satellite)dataset in UCI machine learning library.Experimental results showed that after learning the knowledge of new category,this network could accept the new without
Designing using manufacturing features
Szecsi, T.; Hoque, A. S. M.
2012-04-01
This paper presents a design system that enables the composition of a part using manufacturing features. Features are selected from feature libraries. Upon insertion, the system ensures that the feature does not contradict the design-for-manufacture rules. This helps eliminating costly manufacturing problems. The system is developed as an extension to a commercial CAD/CAM system Pro/Engineer.
刘隽
2008-01-01
Every literature has its features in some aspects,so is the Bible,one of the greatest literary works in the world that has great impact on western literature.This paper summarizes two features of the Bible,namely,cultural feature and literary feature.
Cone inputs to murine striate cortex
Gouras Peter
2008-11-01
Full Text Available Abstract Background We have recorded responses from single neurons in murine visual cortex to determine the effectiveness of the input from the two murine cone photoreceptor mechanisms and whether there is any unique selectivity for cone inputs at this higher region of the visual system that would support the possibility of colour vision in mice. Each eye was stimulated by diffuse light, either 370 (strong stimulus for the ultra-violet (UV cone opsin or 505 nm (exclusively stimulating the middle wavelength sensitive (M cone opsin, obtained from light emitting diodes (LEDs in the presence of a strong adapting light that suppressed the responses of rods. Results Single cells responded to these diffuse stimuli in all areas of striate cortex. Two types of responsive cells were encountered. One type (135/323 – 42% had little to no spontaneous activity and responded at either the on and/or the off phase of the light stimulus with a few impulses often of relatively large amplitude. A second type (166/323 – 51% had spontaneous activity and responded tonically to light stimuli with impulses often of small amplitude. Most of the cells responded similarly to both spectral stimuli. A few (18/323 – 6% responded strongly or exclusively to one or the other spectral stimulus and rarely in a spectrally opponent manner. Conclusion Most cells in murine striate cortex receive excitatory inputs from both UV- and M-cones. A small fraction shows either strong selectivity for one or the other cone mechanism and occasionally cone opponent responses. Cells that could underlie chromatic contrast detection are present but extremely rare in murine striate cortex.
Comparison between Input Hypothesis and Interaction Hypothesis
李佳
2012-01-01
Krashen’s Input hypothesis and Long’s Interaction hypothesis are both valuable research results in the field of language acquisition and play a significant role in language teaching and learning instruction. Through comparing them, their similarities lie in same goal and basis, same focus on comprehension and same challenge the traditional teaching concept. While the differences lie in Different ways to make exposure comprehensible and different roles that learners play. It is meaningful to make the compari⁃son because the results can be valuable guidance and highlights for language teachers and learners to teach or acquire a new lan⁃guage more efficiently.
LISTENING COMPREHENSION: MORE THAN JUST COMPREHENSIBLE INPUT
1994-01-01
In the ten years since the publication of Krashen’s theory on second language acquisition (SLA), the role of comprehensible input (CI) in the learning/acquiring of a language has received considerable attention (Krashen, 1982, 1985; Ellis, 1991, 1992; Long, 1983, 1985). As a result of these studies researchers now agree on the following points. Exposure to a language does not lead to acquisition; the personal accounts of so many language learners who have spent many years in a country or who have listened to endless hours of radio and television without being able to understand or speak the language attest to this fact.
Operational modal analysis with non stationnary inputs
Gouache, Thibault; Morlier, Joseph; Michon, Guilhem; Coulange, Baptiste
2013-01-01
Operational modal analysis (OMA) techniques enable the use of in-situ and uncontrolled vibrations to be used to lead modal analysis of structures. In reality operational vibrations are a combination of numerous excitations sources that are much more complex than a random white noise or a harmonic. Numerous OMA techniques exist like SSI, NExT, FDD and BSS. All these methods are based on the fundamental hypothesis that the input or force applied to the structure to be analyzed is a stationary w...
TSM control of the delayed input system
无
2007-01-01
The paper proposed a terminal sliding mode control method for the delayed input system with uncertainties. Firstly, through the state transformation, the original system was transformed into the non-delayed controllable canonical form system. Then the paper designed a terminal sliding mode and terminal sliding control law with Lyapunov method for the transformed system. Through the method, the reaching time of the any initial state and the convergencing time to the equilibrium points are constrained in finite time. The simulation results show the validation of the method.
Intelligent Graph Layout Using Many Users' Input.
Yuan, Xiaoru; Che, Limei; Hu, Yifan; Zhang, Xin
2012-12-01
In this paper, we propose a new strategy for graph drawing utilizing layouts of many sub-graphs supplied by a large group of people in a crowd sourcing manner. We developed an algorithm based on Laplacian constrained distance embedding to merge subgraphs submitted by different users, while attempting to maintain the topological information of the individual input layouts. To facilitate collection of layouts from many people, a light-weight interactive system has been designed to enable convenient dynamic viewing, modification and traversing between layouts. Compared with other existing graph layout algorithms, our approach can achieve more aesthetic and meaningful layouts with high user preference.
Flexible input, dazzling output with IBM i
Victória-Pereira, Rafael
2014-01-01
Link your IBM i system to the modern business server world! This book presents easier and more flexible ways to get data into your IBM i system, along with rather surprising methods to export and present the vital business data it contains. You'll learn how to automate file transfers, seamlessly connect PC applications with your RPG programs, and much more. Input operations will become more flexible and user-proof, with self-correcting import processes and direct file transfers that require a minimum of user intervention. Also learn novel ways to present information: your DB2 data will look gr
Example of input-output analysis
1975-01-01
The thirty sectors included in the ECASTAR energy input-output model were listed. Five of these belong to energy producing sectors, fifteen to manufacturing industries, two to residential and commercial sectors, and eight to service industries. The model is capable of tracing impacts of an action in three dimensions: dollars, BTU's of energy, and labor. Four conservation actions considered were listed and then discussed separately, dealing with the following areas: increase in fuel efficiency, reduction in fuel used by the transportation and warehousing group, manufacturing of smaller automobiles, and a communications/transportation trade-off.
DeJager, Nathan R.
2017-01-01
The data are input data files to run the forest simulation model Landis-II for Isle Royale National Park. Files include: a) Initial_Comm, which includes the location of each mapcode, b) Cohort_ages, which includes the ages for each tree species-cohort within each mapcode, c) Ecoregions, which consist of different regions of soils and climate, d) Ecoregion_codes, which define the ecoregions, and e) Species_Params, which link the potential establishment and growth rates for each species with each ecoregion.
Culture Input in Foreign Language Teaching
胡晶
2009-01-01
Language and culture are highly interrelated, that is to say, language is not only the carrier of culture but it is also restricted by culture. Therefore, foreign language teaching aiming at cultivate students' intercultural communication should take culture differences into consideration. In this paper, the relationship between language and culture will be discussed. Then I will illustrate the importance of intercultural communication. Finally, according to the present situation of foreign language teaching in China, several strategies for cultural input in and out of class will be suggested.
Approximate input physics for stellar modelling
Pols, O R; Eggleton, P P; Han, Z; Pols, O R; Tout, C A; Eggleton, P P; Han, Z
1995-01-01
We present a simple and efficient, yet reasonably accurate, equation of state, which at the moderately low temperatures and high densities found in the interiors of stars less massive than the Sun is substantially more accurate than its predecessor by Eggleton, Faulkner & Flannery. Along with the most recently available values in tabular form of opacities, neutrino loss rates, and nuclear reaction rates for a selection of the most important reactions, this provides a convenient package of input physics for stellar modelling. We briefly discuss a few results obtained with the updated stellar evolution code.
Apel, Sven; Wendler, Philipp; von Rhein, Alexander; Beyer, Dirk
2011-01-01
A software product line is a set of software products that are distinguished in terms of features (i.e., end-user--visible units of behavior). Feature interactions ---situations in which the combination of features leads to emergent and possibly critical behavior--- are a major source of failures in software product lines. We explore how feature-aware verification can improve the automatic detection of feature interactions in software product lines. Feature-aware verification uses product-line verification techniques and supports the specification of feature properties along with the features in separate and composable units. It integrates the technique of variability encoding to verify a product line without generating and checking a possibly exponential number of feature combinations. We developed the tool suite SPLverifier for feature-aware verification, which is based on standard model-checking technology. We applied it to an e-mail system that incorporates domain knowledge of AT&T. We found that feat...
Unsupervised Feature Subset Selection
Søndberg-Madsen, Nicolaj; Thomsen, C.; Pena, Jose
2003-01-01
This paper studies filter and hybrid filter-wrapper feature subset selection for unsupervised learning (data clustering). We constrain the search for the best feature subset by scoring the dependence of every feature on the rest of the features, conjecturing that these scores discriminate some...... irrelevant features. We report experimental results on artificial and real data for unsupervised learning of naive Bayes models. Both the filter and hybrid approaches perform satisfactorily....
Maleki, Zinat; Pazhakh, AbdolReza
2012-01-01
The present study was an attempt to investigate the effects of premodified input, interactionally modified input and modified output on 80 EFL learners' comprehension of new words. The subjects were randomly assigned into four groups of pre modified input, interactionally modified input, modified output and unmodified (control) groups. Each group…
王冬生; 李世华; 周杏鹏
2011-01-01
针对自来水生产过程的原水水质评价问题,提出了一种基于PSO-RBF神经网络模型的原水水质评价方法.首先,根据水厂生产经验和历史数据分析,制定面向自来水生产过程的原水水质评价标准.然后,采用粒子群优化(PSO)算法训练的RBF神经网络模型,对苏州市相城水厂的进厂原水水质实施在线评价.最后,将进厂原水水质在线评价结果作为前馈量,增加相城水厂药剂(矾和臭氧)投加过程的前馈控制环节,使得药剂投加量能够根据原水水质的变化及时做出调整.实际应用效果表明,与改进前的反馈控制过程相比,过程出水水质更加平稳,提高了自来水生产过程应对原水水质变化的能力.%In consideration of the assessment problem of raw water quality oriented to drinking water treatment process, an assessment method of raw rater quality based on the PSO-RBF neural network model is proposed. First, on the basis of productive experiences and analysis of historical data in the waterworks, an assessment standard oriented to the process of drinking water treatment is established. Then, the radial basis function (RBF) neural network model trained by the particle swarm optimization ( PSO) algorithm is used for the on-line assessment of raw water quality in the Xiangcheng Waterworks of Suzhou city. Finally, feed-forward control elements are added to the pharmaceutical (alum and ozone) dosing control processes of Xiangcheng Waterworks, using the online assessment result as the feed-forward compensation. The results of the practical operation show that the produced water quality becomes more stable, and the adaptation ability of drinking water treatment to the variation of raw water quality is improved.
B. Balakumar
2013-09-01
Full Text Available Magnetic Resonance Images (MRI are widely used in the diagnosis of Brain tumor. In this study we have developed a new approach for automatic classification of the normal and abnormal non-enhanced MRI images. The proposed method consists of four stages namely Preprocessing, feature extraction, feature reduction and classification. In the first stage anisotropic filter is applied for noise reduction and to make the image suitable for extracting the features. In the second stage, Region growing base segmentation is used for partitioning the image into meaningful regions. In the third stage, combined edge and Texture based features are extracted using Histogram and Gray Level Co-occurrence Matrix (GLCM from the segmented image. In the next stage PCA is used to reduce the dimensionality of the Feature space which results in a more efficient and accurate classification. Finally, in the classification stage, a supervised Radial Basics Function (RBF classifier is used to classify the experimental images into normal and abnormal. The obtained experimental are evaluated using the metrics sensitivity, specificity and accuracy. For comparison, the performance of the proposed technique has significantly improved the tumor detection accuracy with other neural network based classifier SVM, FFNN and FSVM.
Cometary micrometeorites and input of prebiotic compounds
Engrand C.
2014-02-01
Full Text Available The apparition of life on the early Earth was probably favored by inputs of extraterrestrial matter brought by carbonaceous chondrite-like objects or cometary material. Interplanetary dust collected nowadays on Earth is related to carbonaceous chondrites and to cometary material. They contain in particular at least a few percent of organic matter, organic compounds (amino-acids, PAHs,…, hydrous silicates, and could have largely contributed to the budget of prebiotic matter on Earth, about 4 Ga ago. A new population of cometary dust was recently discovered in the Concordia Antarctic micrometeorite collection. These “Ultracarbonaceous Antarctic Micrometeorites” (UCAMMs are dominated by deuterium-rich and nitrogen-rich organic matter. They seem related to the “CHON” grains identified in the comet Halley in 1986. Although rare in the micrometeorites flux (<5% of the micrometeorites, UCAMMs could have significantly contributed to the input of prebiotic matter. Their content in soluble organic matter is currently under study.
Processing in (linear) systems with stochastic input
Nutu, Catalin Silviu; Axinte, Tiberiu
2016-12-01
The paper is providing a different approach to real-world systems, such as micro and macro systems of our real life, where the man has little or no influence on the system, either not knowing the rules of the respective system or not knowing the input of the system, being thus mainly only spectator of the system's output. In such a system, the input of the system and the laws ruling the system could be only "guessed", based on intuition or previous knowledge of the analyzer of the respective system. But, as we will see in the paper, it exists also another, more theoretical and hence scientific way to approach the matter of the real-world systems, and this approach is mostly based on the theory related to Schrödinger's equation and the wave function associated with it and quantum mechanics as well. The main results of the paper are regarding the utilization of the Schrödinger's equation and related theory but also of the Quantum mechanics, in modeling real-life and real-world systems.
Ground motion input in seismic evaluation studies
Sewell, R.T.; Wu, S.C.
1996-07-01
This report documents research pertaining to conservatism and variability in seismic risk estimates. Specifically, it examines whether or not artificial motions produce unrealistic evaluation demands, i.e., demands significantly inconsistent with those expected from real earthquake motions. To study these issues, two types of artificial motions are considered: (a) motions with smooth response spectra, and (b) motions with realistic variations in spectral amplitude across vibration frequency. For both types of artificial motion, time histories are generated to match target spectral shapes. For comparison, empirical motions representative of those that might result from strong earthquakes in the Eastern U.S. are also considered. The study findings suggest that artificial motions resulting from typical simulation approaches (aimed at matching a given target spectrum) are generally adequate and appropriate in representing the peak-response demands that may be induced in linear structures and equipment responding to real earthquake motions. Also, given similar input Fourier energies at high-frequencies, levels of input Fourier energy at low frequencies observed for artificial motions are substantially similar to those levels noted in real earthquake motions. In addition, the study reveals specific problems resulting from the application of Western U.S. type motions for seismic evaluation of Eastern U.S. nuclear power plants.
Image inputs in Structure-from-Motion Photogrammetry: optimising image greyscaling
O'Connor, James; Smith, Mike J.; James, Mike R.
2016-04-01
Structure-from-motion (SfM) photogrammetry is an emerging technology receiving much attention within the geoscience community due to its ease of use and the lack of prior information required to build topographic models from images. However, little consideration is given to image inputs considering image sharpness and contrast both have an effect on the quality of photogrammetric outputs. This task is made more challenging across natural image sequences due to the presence of low-contrast surfaces which are often at oblique angles to input images. As most feature detectors operate on a single image channel, monochrome inputs can be pre-processed for input into SfM workflows and relative accuracy measured. In this contribution we process two sets of imagery from both a real world, close range scenario (Constitution Hill, Aberystwyth) and a controlled dataset in laboratory conditions simulating a UAV flight with convergent viewing geometry. With each, we generate greyscale subsets comprised of weighted combinations of the spectral bands of the input images prior to executing SfM workflows. Output point clouds are measured against high-accuracy terrestrial laser scans in order to assess residual error and compare output solutions. When compared with untreated image inputs into a commonly used commercial package (Agisoft Photoscan Pro) we show minor improvements in the accuracy of photogrammetrically derived products.
Supplementary High-Input Impedance Voltage-Mode Universal Biquadratic Filter Using DVCCs
Jitendra Mohan
2012-01-01
Full Text Available To further extend the existing knowledge on voltage-mode universal biquadratic filter, in this paper, a new biquadratic filter circuit with single input and multiple outputs is proposed, employing three differential voltage current conveyors (DVCCs, three resistors, and two grounded capacitors. The proposed circuit realizes all the standard filter functions, that is, high-pass, band-pass, low-pass, notch, and all-pass filters simultaneously. The circuit enjoys the feature of high-input impedance, orthogonal control of resonance angular frequency (o, and quality factor (Q via grounded resistor and the use of grounded capacitors which is ideal for IC implementation.
Multiple-Input Multiple-Output (MIMO Linear Systems Extreme Inputs/Outputs
David O. Smallwood
2007-01-01
Full Text Available A linear structure is excited at multiple points with a stationary normal random process. The response of the structure is measured at multiple outputs. If the autospectral densities of the inputs are specified, the phase relationships between the inputs are derived that will minimize or maximize the trace of the autospectral density matrix of the outputs. If the autospectral densities of the outputs are specified, the phase relationships between the outputs that will minimize or maximize the trace of the input autospectral density matrix are derived. It is shown that other phase relationships and ordinary coherence less than one will result in a trace intermediate between these extremes. Least favorable response and some classes of critical response are special cases of the development. It is shown that the derivation for stationary random waveforms can also be applied to nonstationary random, transients, and deterministic waveforms.
Design Features of Modern Mechanical Ventilators.
MacIntyre, Neil
2016-12-01
A positive-pressure breath ideally should provide a VT that is adequate for gas exchange and appropriate muscle unloading while minimizing any risk for injury or discomfort. The latest generation of ventilators uses sophisticated feedback systems to sculpt positive-pressure breaths according to patient effort and respiratory system mechanics. Currently, however, these new control strategies are not totally closed-loop systems. This is because the automatic input variables remain limited, some clinician settings are still required, and the specific features of the perfect breath design still are not entirely clear. Despite these limitations, there are some rationale for many of these newer feedback features.
REVERSE MODELING FOR CONIC BLENDING FEATURE
Fan Shuqian; Ke Yinglin
2005-01-01
A novel method to extract conic blending feature in reverse engineering is presented.Different from the methods to recover constant and variable radius blends from unorganized points, it contains not only novel segmentation and feature recognition techniques, but also bias corrected technique to capture more reliable distribution of feature parameters along the spine curve. The segmentation depending on point classification separates the points in the conic blend region from the input point cloud. The available feature parameters of the cross-sectional curves are extracted with the processes of slicing point clouds with planes, conic curve fitting, and parameters estimation and compensation. The extracted parameters and its distribution laws are refined according to statistic theory such as regression analysis and hypothesis test. The proposed method can accurately capture the original design intentions and conveniently guide the reverse modeling process. Application examples are presented to verify the high precision and stability of the proposed method.
Analysis on relation between safety input and accidents
YAO Qing-guo; ZHANG Xue-mu; LI Chun-hui
2007-01-01
The number of safety input directly determines the level of safety, and there exists dialectical and unified relations between safety input and accidents. Based on the field investigation and reliable data, this paper deeply studied the dialectical relationship between safety input and accidents, and acquired the conclusions. The security situation of the coal enterprises was related to the security input rate, being effected little by the security input scale, and build the relationship model between safety input and accidents on this basis, that is the accident model.
Auto Draw from Excel Input Files
Strauss, Karl F.; Goullioud, Renaud; Cox, Brian; Grimes, James M.
2011-01-01
The design process often involves the use of Excel files during project development. To facilitate communications of the information in the Excel files, drawings are often generated. During the design process, the Excel files are updated often to reflect new input. The problem is that the drawings often lag the updates, often leading to confusion of the current state of the design. The use of this program allows visualization of complex data in a format that is more easily understandable than pages of numbers. Because the graphical output can be updated automatically, the manual labor of diagram drawing can be eliminated. The more frequent update of system diagrams can reduce confusion and reduce errors and is likely to uncover symmetric problems earlier in the design cycle, thus reducing rework and redesign.
刘述文; 潘宏侠; 刘涛涛
2015-01-01
局域均值分解(Local Mean Decomposition,LMD)是近年来出现的一种新的时频分析方法,在机械设备故障诊断领域中的应用日益广泛.针对齿轮箱振动故障信号的非平稳性和非线性,提出了一种基于局域均值分解和径向基函数神经网络(Radial BasisFunction Neural Network,RBF)相结合的齿轮箱故障诊断方法.该方法利用小波包对原始信号进行消噪;利用LMD对处理后信号进行分解,得到一系列PF分量(Product Function,PF);选取包含主要故障信息的PF分量并从中提取偏度系数等特征参数对RBF神经网络进行训练,并对齿轮箱故障进行识别和分类.通过实例验证了该方法的有效性.
胡浩; 何子辉
2012-01-01
Based on the analysis of induction motor loss model, and aimed at the complicated nonlinear relation between the torque and rotational speed with optimal excitation, this paper puts forward a RBF neural network and establishes the model of induction motor efficiency optimization. It ues the MATLAB to simulate the system. The experimental results show the system＇s efficiency is improved and the loss of induction motor is reduced significantly.%在对磁链定向下感应电机损耗模型进行了详细的分析基础上，针对电机转矩和转速与最优励磁电流存在严重的非线性关系，文中提出一种径向基神经网络控制方法并建立电机效率优化控制模型，对电机进行最大效率优化控制。仿真结果表明该系统运行效率明显提高，降低了电机损耗。
郭兰平; 俞建宁; 张建刚; 漆玉娟; 张旭东
2011-01-01
提出一种基于遗传算法和RBF神经网络相结合的时间序列预测模型,克服了单个神经网络在非线性时间序列预测中容易陷入局部极小值及网络训练速度缓慢的问题.以居民消费价格指数数据进行训练和测试,与传统的BP神经网络预测模型相比较,该模型的预测精度是令人满意的,数值模拟证明了该方法的有效性和可行性.%A time series forecasting model based on genetic algorithm and RBF neural network is proposed.The problem that single neural network in nonlinear time series forecasting easily gets into the local minimum and neural network has a very slow study rate is overcome. The model is then used to forecast the inhabitant consumer price index (CPI). Compared with the traditional BP neural network forecast model, this model forecast precision is satisfying. Numerical simulation illustrates the feasibility of the technique.
蔡时连; 许亮
2012-01-01
Book circulation is an important index for estimating of library.To resolve the forecast problem of book circulation,BP and RBF neural network predictive models are introduced in this paper.Based on the book circulation data of architectural book in the library from 2002 to 2010,Matlab simulation is implemented.The results show that the book circulation can be predicted by the two models effectively.%图书流通量是评价图书馆工作的重要指标.为解决图书流通量预测问题,引入BP和RBF神经网络预测模型.结合北京建筑工程学院图书馆2002年至2010年建筑类图书的流通数据进行了matlab仿真,结果表明,两种模型都能有效预测图书的流通量.
基于RBF神经网络的军事物流设施规模预测%Prediction of Military Logistics Facilities Scale Based on RBF Neural Network
刘洪娟; 甘明; 肖育; 姜玉宏
2013-01-01
确定合理的军事物流设施规模对于成功实施军事后勤保障十分重要。提出了基于RBF神经网络的军事物流设施规模预测模型的建模方法，该方法旨在确定军事物流设施规模与其影响因素之间的非线性关系；采用算例说明了基于RBF神经网络的军事物流设施规模预测的具体做法，对军事物流设施规模的确定具有指导意义。%It is important for successfully implementing the military logistics support to determine the reasonable facilities scale of military logistics. This paper puts forward the modeling method for prediction of military logistics facilities scale based on RBF neural network. The purpose of this method is for determining the nonlinear relationship between military logistics facilities scale and its related influence factors. Detailed operations of ascertaining facilities scale of military logistics have been illustrated by usage of an example. This method is a guidance for determining the military logistics facilities scale.
Pickling Process Model of Zr-4 Alloy Tube Based on RBF Network%基于RBF神经网络法的Zr-4合金管材酸洗工艺模型
卫新民; 袁改焕; 李小宁
2015-01-01
研究了Zr-4合金管材酸洗处理过程中,酸洗去除量、酸水转换时间、冲水时间及酸洗次数对管材氟残留量的影响,并基于径向基(RBF)人工神经网络法建立了Zr4合金管材酸洗工艺与氟残留的神经网络模型.结果表明:冲水酸水转换时间和冲水时间对氟残留量均有影响,且酸水转换时间的影响更为显著;氟残留量与酸洗次数无明显对应关系.Zr4合金酸洗工艺的RBF神经网络模型结构为3-5-1,实际值与模拟值的相对误差为9.2％.该神经网络模型具有较高的可靠性,可为Zr-4合金酸洗工艺参数的优化提供参考.
王蕾
2014-01-01
Because of the construction of logistics park in our country is in operation of the site planning and management need to explore in practice ,therefore,this article in view of the logistics park location selection problem ,to build the site selection evaluation system model based on RBF neural network to applied research ,will be for our country's logistics park planning and development provides a new theoretical basis and the overall framework of reference .%由于我国物流园区的建设在选址规划和管理运作方面还需在实际中探索，因此，文中针对物流园区选址问题，构建基于RBF神经网络的选址评价体系模型进行应用研究，将为我国的物流园区规划发展提供了新的理论基础和整体框架参考。
张雷; 胡彦红; 陈巍巍; 刘秋鞍; 林建中; 张丽芳
2010-01-01
在径向基函数(Radial Basis Function,RBF)神经网络成熟的基础上,对旋转机械的转子系统进行故障诊断,针对梯度下降法容易产生梯度消失的问题,提出用扩展卡尔曼滤波器(Extended Kalman Filter,EKF)对权重进行调节训练,并将结果与反向传播(Back Propagation,BP)算法和梯度下降调节进行比较,用EKF训练的RBF神经网络不仅在性能上有优势,在精度和迭代速度上亦优于其他方法.相信在今后的实际应用中尤其在旋转机械故障诊断中可以更大地发挥其优势.
Study on Rear Axle Gear Residual Life Predication based on RBF Network%基于RBF网络的后桥齿轮残余寿命预测研究
曾宇露; 祝志芳
2012-01-01
为有效预测汽车后桥齿轮的残余寿命,针对后桥的非线性特性,提出了一种递归预处理与RBF网络相结合的齿轮残余寿命预测方法,并验证了该方法的可行性.利用该方法进行了汽车后桥齿轮残余寿命预测,结果表明,该方法对齿轮残余寿命的预测结果与齿轮疲劳试验结果吻合,预测精度高.%For the non-linearity property of rear axle, a new method, which consists of a recursive pretreatment and RBF neural networks, is presented in this paper to accurately predict residual life of rear axle gear. Simulation and experiment results have proved feasibility of this method. This method can be used to predict the residual life of rear axle gear, the results show that predication of gear residual life with this method is very accurate, and consistent with gear fatigue test results.
Distribution Development for STORM Ingestion Input Parameters
Fulton, John [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
2017-07-01
The Sandia-developed Transport of Radioactive Materials (STORM) code suite is used as part of the Radioisotope Power System Launch Safety (RPSLS) program to perform statistical modeling of the consequences due to release of radioactive material given a launch accident. As part of this modeling, STORM samples input parameters from probability distributions with some parameters treated as constants. This report described the work done to convert four of these constant inputs (Consumption Rate, Average Crop Yield, Cropland to Landuse Database Ratio, and Crop Uptake Factor) to sampled values. Consumption rate changed from a constant value of 557.68 kg / yr to a normal distribution with a mean of 102.96 kg / yr and a standard deviation of 2.65 kg / yr. Meanwhile, Average Crop Yield changed from a constant value of 3.783 kg edible / m ^{2} to a normal distribution with a mean of 3.23 kg edible / m ^{2} and a standard deviation of 0.442 kg edible / m ^{2} . The Cropland to Landuse Database ratio changed from a constant value of 0.0996 (9.96%) to a normal distribution with a mean value of 0.0312 (3.12%) and a standard deviation of 0.00292 (0.29%). Finally the crop uptake factor changed from a constant value of 6.37e^{-4} (Bq crop /kg)/(Bq soil /kg) to a lognormal distribution with a geometric mean value of 3.38e^{-4} (Bq crop /kg)/(Bq soil /kg) and a standard deviation value of 3.33 (Bq crop /kg)/(Bq soil /kg)
蔡坤; 刘兴高
2012-01-01
内部热耦合精馏(ITCDIC)是迄今为止所提出的四大精馏节能技术中节能效果最高,但唯一没有商业化的节能技术,比常规精馏要节能40％以上,没有商业化的主要种类基于减聚类、K-means原因之一在于该过程具有较强的非线性、复杂动态特性以及耦合性,给控制方案的设计带来了诸多困难.由于径向基(RBF)神经网络具有快速学习并能逼近任意非线性函数的优点,本文提出了一种基于RBF神经网络内模控制的混合优化算法,是一种粒子群优化的混合优化算法,以苯-甲苯物系作为研究实例,并与国际公开报道的结果进行了详细比较,研究结果表明基于混合优化算法的RBF神经网络内模控制相比于传统的PID、常规RBF算法和国际公开报道有着更好的控制效果.%Internal thermally coupled distillation is the most promising of four major distillation energy-saving technologies, which can save more than 40 % energy compared with traditional distillation process, but it has not been widely used. The bottleneck that prevents the process from being commercialized is the operational difficulties due to the nonlinearity, complex dynamics and interactive nature of the process. The radial basis(RBF) neural network has fast learning and can identify any nonlinear function. We presented a hybrid optimization algorithm for RBF neural network, which is based on particle swarm optimization, gradient descent method, K-means clustering and subtractive clustering algorithm. Take the benzene-toluene system as a research case and compared with the international public reporting results. The hybrid optimization algorithm for RBF neural network is more reliable than PID algorithm, conventional RBF and the international public reporting results.
Consumer input into research: the Australian Cancer Trials website
Butow Phyllis N
2011-06-01
Full Text Available Abstract Background The Australian Cancer Trials website (ACTO was publicly launched in 2010 to help people search for cancer clinical trials recruiting in Australia, provide information about clinical trials and assist with doctor-patient communication about trials. We describe consumer involvement in the design and development of ACTO and report our preliminary patient evaluation of the website. Methods Consumers, led by Cancer Voices NSW, provided the impetus to develop the website. Consumer representative groups were consulted by the research team during the design and development of ACTO which combines a search engine, trial details, general information about trial participation and question prompt lists. Website use was analysed. A patient evaluation questionnaire was completed at one hospital, one week after exposure to the website. Results ACTO's main features and content reflect consumer input. In February 2011, it covered 1, 042 cancer trials. Since ACTO's public launch in November 2010, until the end of February 2011, the website has had 2, 549 new visits and generated 17, 833 page views. In a sub-study of 47 patient users, 89% found the website helpful for learning about clinical trials and all respondents thought patients should have access to ACTO. Conclusions The development of ACTO is an example of consumers working with doctors, researchers and policy makers to improve the information available to people whose lives are affected by cancer and to help them participate in their treatment decisions, including consideration of clinical trial enrolment. Consumer input has ensured that the website is informative, targets consumer priorities and is user-friendly. ACTO serves as a model for other health conditions.
Robust input design for nonlinear dynamic modeling of AUV.
Nouri, Nowrouz Mohammad; Valadi, Mehrdad
2017-09-01
Input design has a dominant role in developing the dynamic model of autonomous underwater vehicles (AUVs) through system identification. Optimal input design is the process of generating informative inputs that can be used to generate the good quality dynamic model of AUVs. In a problem with optimal input design, the desired input signal depends on the unknown system which is intended to be identified. In this paper, the input design approach which is robust to uncertainties in model parameters is used. The Bayesian robust design strategy is applied to design input signals for dynamic modeling of AUVs. The employed approach can design multiple inputs and apply constraints on an AUV system's inputs and outputs. Particle swarm optimization (PSO) is employed to solve the constraint robust optimization problem. The presented algorithm is used for designing the input signals for an AUV, and the estimate obtained by robust input design is compared with that of the optimal input design. According to the results, proposed input design can satisfy both robustness of constraints and optimality. Copyright © 2017 ISA. Published by Elsevier Ltd. All rights reserved.
White, Lydia
1987-01-01
Discusses several objections to Krashen's Input Hypothesis which states that language acquisition is the learners' understanding of a language at a stage slightly higher than their current one because of their understanding of extralinguistic cues of the language. (Author/LMO)
Miguel Kiguel
1995-03-01
Full Text Available Fischer's classic (1974 paper develops conditions under which it is appropriate to use money as an input in a 'delivered' production function. In this paper, we extend Fischer's model I (the Baumol-Tobin inventory approach by incorporating credit into the analysis. Our investigation of the extended model brings out a very restrictive but necessary implicit assumption employed by Fischer to treat money as an input. Namely. that there exists a binding constraint on the use of money! A similar result holds for our more general model. Fischer's classic (1974 paper develops conditions under which it is appropriate to use money as an input in a 'delivered' production function. In this paper, we extend Fischer's model I (the Baumol-Tobin inventory approach by incorporating credit into the analysis. Our investigation of the extended model brings out a very restrictive but necessary implicit assumption employed by Fischer to treat money as an input. Namely. that there exists a binding constraint on the use of money! A similar result holds for our more general model.
Hybrid input function estimation using a single-input-multiple-output (SIMO) approach
Su, Yi; Shoghi, Kooresh I.
2009-02-01
A hybrid blood input function (BIF) model that incorporates region of interests (ROIs) based peak estimation and a two exponential tail model was proposed to describe the blood input function. The hybrid BIF model was applied to the single-input-multiple-output (SIMO) optimization based approach for BIF estimation using time activity curves (TACs) obtained from ROIs defined at left ventricle (LV) blood pool and myocardium regions of dynamic PET images. The proposed BIF estimation method was applied with 0, 1 and 2 blood samples as constraints for BIF estimation using simulated small animal PET data. Relative percentage difference of the area-under-curve (AUC) measurement between the estimated BIF and the true BIF was calculated to evaluate the BIF estimation accuracy. SIMO based BIF estimation using Feng's input function model was also applied for comparison. The hybrid method provided improved BIF estimation in terms of both mean accuracy and variability compared to Feng's model based BIF estimation in our simulation study. When two blood samples were used as constraints, the percentage BIF estimation error was 0.82 +/- 4.32% for the hybrid approach and 4.63 +/- 10.67% for the Feng's model based approach. Using hybrid BIF, improved kinetic parameter estimation was also obtained.
Halliwell, Emily R; Jones, Linor L; Fraser, Matthew; Lockley, Morag; Hill-Feltham, Penelope; McKay, Colette M
2015-06-01
A study was conducted to determine whether modifications to input compression and input frequency response characteristics can improve music-listening satisfaction in cochlear implant users. Experiment 1 compared three pre-processed versions of music and speech stimuli in a laboratory setting: original, compressed, and flattened frequency response. Music excerpts comprised three music genres (classical, country, and jazz), and a running speech excerpt was compared. Experiment 2 implemented a flattened input frequency response in the speech processor program. In a take-home trial, participants compared unaltered and flattened frequency responses. Ten and twelve adult Nucleus Freedom cochlear implant users participated in Experiments 1 and 2, respectively. Experiment 1 revealed a significant preference for music stimuli with a flattened frequency response compared to both original and compressed stimuli, whereas there was a significant preference for the original (rising) frequency response for speech stimuli. Experiment 2 revealed no significant mean preference for the flattened frequency response, with 9 of 11 subjects preferring the rising frequency response. Input compression did not alter music enjoyment. Comparison of the two experiments indicated that individual frequency response preferences may depend on the genre or familiarity, and particularly whether the music contained lyrics.
Meakins, Felicity; Wigglesworth, Gillian
2013-01-01
In situations of language endangerment, the ability to understand a language tends to persevere longer than the ability to speak it. As a result, the possibility of language revival remains high even when few speakers remain. Nonetheless, this potential requires that those with high levels of comprehension received sufficient input as children for…
Shahrukh Adnan Khan M. D.
2017-01-01
Full Text Available This paper presents a Graphical User Interface (GUI software utility for the input/output characterization of single variable and multivariable nonlinear systems by obtaining the sinusoidal input describing function (SIDF of the plant. The software utility is developed on MATLAB R2011a environment. The developed GUI holds no restriction on the nonlinearity type, arrangement and system order; provided that output(s of the system is obtainable either though simulation or experiments. An insight to the GUI and its features are presented in this paper and example problems from both single variable and multivariable cases are demonstrated. The formulation of input/output behavior of the system is discussed and the nucleus of the MATLAB command underlying the user interface has been outlined. Some of the industries that would benefit from this software utility includes but not limited to aerospace, defense technology, robotics and automotive.
Starikov, Sergey N.; Konnik, Mikhail V.; Manykin, Edward A.; Rodin, Vladislav G.
2009-04-01
Linear methods of restoration of input scene's images in optical-digital correlators are described. Relatively low signal to noise ratio of a camera's photo sensor and extensional PSF's size are special features of considered optical-digital correlator. RAW-files of real correlation signals obtained by digital photo sensor were used for input scene's images restoration. It is shown that modified evolution method, which employs regularization by Tikhonov, is better among linear deconvolution methods. As a regularization term, an inverse signal to noise ratio as a function of spatial frequencies was used. For additional improvement of restoration's quality, noise analysis of boundary areas of the image to be reconstructed was performed. Experimental results on digital restoration of input scene's images are presented.
Ambiguities in input-output behavior of driven nonlinear systems close to bifurcation
Reit Marco
2016-06-01
Full Text Available Since the so-called Hopf-type amplifier has become an established element in the modeling of the mammalian hearing organ, it also gets attention in the design of nonlinear amplifiers for technical applications. Due to its pure sinusoidal response to a sinusoidal input signal, the amplifier based on the normal form of the Andronov-Hopf bifurcation is a peculiar exception of nonlinear amplifiers. This feature allows an exact mathematical formulation of the input-output characteristic and thus deeper insights of the nonlinear behavior. Aside from the Hopf-type amplifier we investigate an extension of the Hopf system with focus on ambiguities, especially the separation of solution sets, and double hysteresis behavior in the input-output characteristic. Our results are validated by a DSP implementation.
A Study on the Input Hypothesis and Interaction Hypothesis
李雪清
2016-01-01
In Second Language Acquisition theory, input and interaction are considered as two key factors greatly influencing the learners’acquisition rate and quality, and therefore input and interaction research has been receiving increasing attention in re-cent years. Among the large amount of research, Krashen’s input hypothesis and Long’s interaction hypothesis are perhaps most influential theories, from which most of input and interaction studies have developed. Input hypothesis claims that compre-hensible input is the only one way to acquire language, whereas interaction hypothesis argues that interaction is necessary for language acquisition. Therefore,this thesis attempts to conduct a descriptive analysis between input hypothesis and interaction hypothesis, based on their basic ideas, theoretical basis, comparisons and empirical work. It concludes that input hypothesis and interaction hypothesis succeed in interpreting the process of language acquisition to some extent, and offer both theoretical and practical inspirations on second language teaching.
Characterization of Input Current Interharmonics in Adjustable Speed Drives
Soltani, Hamid; Davari, Pooya; Zare, Firuz
2017-01-01
-edge symmetrical regularly sampled Space Vector Modulation (SVM) technique, on the input current interharmonic components are presented and discussed. Particular attention is also given to the influence of the asymmetrical regularly sampled modulation technique on the drive input current interharmonics...
STABILITY ANALYSIS OF THE DYNAMIC INPUT-OUTPUT SYSTEM
GuoChonghui; TangHuanwen
2002-01-01
The dynamic input-output model is well known in economic theory and practice. In this paper, the asymptotic stability and balanced growth solutions of the dynamic input-output system are considered. Under some natural assumptions which do not require the technical coefficient matrix to be indecomposable,it has been proved that the dynamic input-output system is not asymptotically stable and the closed dynamic input-output model has a balanced growth solution.
Estimating nonstationary input signals from a single neuronal spike train
Kim, Hideaki; Shinomoto, Shigeru
2012-01-01
Neurons temporally integrate input signals, translating them into timed output spikes. Because neurons nonperiodically emit spikes, examining spike timing can reveal information about input signals, which are determined by activities in the populations of excitatory and inhibitory presynaptic neurons. Although a number of mathematical methods have been developed to estimate such input parameters as the mean and fluctuation of the input current, these techniques are based on the unrealistic as...
Input modelling for subchannel analysis of CANFLEX fuel bundle
Park, Joo Hwan; Jun, Ji Su; Suk, Ho Chun [Korea Atomic Energy Research Institute, Taejon (Korea)
1998-06-01
This report describs the input modelling for subchannel analysis of CANFLEX fuel bundle using CASS(Candu thermalhydraulic Analysis by Subchannel approacheS) code which has been developed for subchannel analysis of CANDU fuel channel. CASS code can give the different calculation results according to users' input modelling. Hence, the objective of this report provide the background information of input modelling, the accuracy of input data and gives the confidence of calculation results. (author). 11 refs., 3 figs., 4 tabs.
Regional Input Output Table for the State of Punjab
Singh, Inderjeet; Singh, Lakhwinder
2011-01-01
Because of policy relevance of regional input-output analysis, a vast literature on the construction of regional input-output tables has emerged in the recent past, especially on the non-survey and hybrid methods. Although, construction of regional input-output tables is not new in India, but generation of input-output table using non-survey methods is relatively a rare phenomenon. This work validates alternative non-survey, location quotient methodologies and finally uses comparatively bette...
Parallel Feature Extraction System
MAHuimin; WANGYan
2003-01-01
Very high speed image processing is needed in some application specially for weapon. In this paper, a high speed image feature extraction system with parallel structure was implemented by Complex programmable logic device (CPLD), and it can realize image feature extraction in several microseconds almost with no delay. This system design is presented by an application instance of flying plane, whose infrared image includes two kinds of feature: geometric shape feature in the binary image and temperature-feature in the gray image. Accordingly the feature extraction is taken on the two kind features. Edge and area are two most important features of the image. Angle often exists in the connection of the different parts of the target's image, which indicates that one area ends and the other area begins. The three key features can form the whole presentation of an image. So this parallel feature extraction system includes three processing modules: edge extraction, angle extraction and area extraction. The parallel structure is realized by a group of processors, every detector is followed by one route of processor, every route has the same circuit form, and works together at the same time controlled by a set of clock to realize feature extraction. The extraction system has simple structure, small volume, high speed, and better stability against noise. It can be used in the war field recognition system.
42 CFR 460.138 - Committees with community input.
2010-10-01
... 42 Public Health 4 2010-10-01 2010-10-01 false Committees with community input. 460.138 Section 460.138 Public Health CENTERS FOR MEDICARE & MEDICAID SERVICES, DEPARTMENT OF HEALTH AND HUMAN... community input. A PACE organization must establish one or more committees, with community input, to do...
Comparison of Linear Microinstability Calculations of Varying Input Realism
G. Rewoldt
2003-09-08
The effect of varying ''input realism'' or varying completeness of the input data for linear microinstability calculations, in particular on the critical value of the ion temperature gradient for the ion temperature gradient mode, is investigated using gyrokinetic and gyrofluid approaches. The calculations show that varying input realism can have a substantial quantitative effect on the results.
On the Nature of the Input in Optimality Theory
Heck, Fabian; Müller, Gereon; Vogel, Ralf;
2002-01-01
The input has two main functions in optimality theory (Prince and Smolensky 1993). First, the input defines the candidate set, in other words it determines which output candidates compete for optimality, and which do not. Second, the input is referred to by faithfulness constraints that prohibit...
REFLECTIONS ON THE INOPERABILITY INPUT-OUTPUT MODEL
Dietzenbacher, Erik; Miller, Ronald E.
2015-01-01
We argue that the inoperability input-output model is a straightforward - albeit potentially very relevant - application of the standard input-output model. In addition, we propose two less standard input-output approaches as alternatives to take into consideration when analyzing the effects of disa
Botma, J.H.; Wassenaar, R.F.; Wiegerink, R.J.
1993-01-01
In this paper a low-voltage two-stage Op Amp is presented. The Op Amp features rail-to-rail operation and has an @put stage with a constant transconductance (%) over the entire common-mode input range. The input stage consists of an n- and a PMOS differential pair connected in parallel. The constant
High-rate laser metal deposition of Inconel 718 component using low heat-input approach
Kong, C. Y.; Scudamore, R. J.; Allen, J.
Currently many aircraft and aero engine components are machined from billets or oversize forgings. This involves significant cost, material wastage, lead-times and environmental impacts. Methods to add complex features to another component or net-shape surface would offer a substantial cost benefit. Laser Metal Deposition (LMD), currently being applied to the repair of worn or damaged aero engine components, was attempted in this work as an alternative process route, to build features onto a base component, because of its low heat input capability. In this work, low heat input and high-rate deposition was developed to deposit Inconel 718 powder onto thin plates. Using the optimised process parameters, a number of demonstrator components were successfully fabricated.
Comparison between Input Hypothesis and Interaction Hypothesis
宗琦
2016-01-01
Second Language Acquisition has received more and more attention since 1950s when it becomes an autonomous field of research. Linguists have carried out many theoretical and empirical studies with a sharp purpose to promote Second Language Acquisition. Krashen’s Input Hypothesis and Long’s Interaction Hypothesis are most influential ones among the studies. They both play important roles in language teaching and learning. The paper will present an account of the two great theories, includ-ing the main claims, theoretical foundations as well as some related empirical works and try to investigate commons and differ-ences between them, based on literature and empirical studies. The purpose of writing this paper is to provide a clear outline of the two theories and point out how they are interrelated yet separate predictions about how second language are learned. It is meaningful because the results can be valuable guidance and highlights for language teachers and learners to teach or acquire a language better.
Investigating Text Input Methods for Mobile Phones
Barry ORiordan
2005-01-01
Full Text Available Human Computer Interaction is a primary factor in the success or failure of any device but if an objective view is taken of the current mobile phone market you would be forgiven for thinking usability was secondary to aesthetics. Many phone manufacturers modify the design of phones to be different than the competition and to target fashion trends, usually at the expense of usability and performance. There is a lack of awareness among many buyers of the usability of the device they are purchasing and the disposability of modern technology is an effect rather than a cause of this. Designing new text entry methods for mobile devices can be expensive and labour-intensive. The assessment and comparison of a new text entry method with current methods is a necessary part of the design process. The best way to do this is through an empirical evaluation. The aim of the study was to establish which mobile phone text input method best suits the requirements of a select group of target users. This study used a diverse range of users to compare devices that are in everyday use by most of the adult population. The proliferation of the devices is as yet unmatched by the study of their application and the consideration of their user friendliness.
Kameda, Hiroshi; Hioki, Hiroyuki; Tanaka, Yasuyo H; Tanaka, Takuma; Sohn, Jaerin; Sonomura, Takahiro; Furuta, Takahiro; Fujiyama, Fumino; Kaneko, Takeshi
2012-03-01
To examine inputs to parvalbumin (PV)-producing interneurons, we generated transgenic mice expressing somatodendritic membrane-targeted green fluorescent protein specifically in the interneurons, and completely visualized their dendrites and somata. Using immunolabeling for vesicular glutamate transporter (VGluT)1, VGluT2, and vesicular GABA transporter, we found that VGluT1-positive terminals made contacts 4- and 3.1-fold more frequently with PV-producing interneurons than VGluT2-positive and GABAergic terminals, respectively, in the primary somatosensory cortex. Even in layer 4, where VGluT2-positive terminals were most densely distributed, VGluT1-positive inputs to PV-producing interneurons were 2.4-fold more frequent than VGluT2-positive inputs. Furthermore, although GABAergic inputs to PV-producing interneurons were as numerous as VGluT2-positive inputs in most cortical layers, GABAergic inputs clearly preferred the proximal dendrites and somata of the interneurons, indicating that the sites of GABAergic inputs were more optimized than those of VGluT2-positive inputs. Simulation analysis with a PV-producing interneuron model compatible with the present morphological data revealed a plausible reason for this observation, by showing that GABAergic and glutamatergic postsynaptic potentials evoked by inputs to distal dendrites were attenuated to 60 and 87%, respectively, of those evoked by somatic inputs. As VGluT1-positive and VGluT2-positive axon terminals were presumed to be cortical and thalamic glutamatergic inputs, respectively, cortical excitatory inputs to PV-producing interneurons outnumbered the thalamic excitatory and intrinsic inhibitory inputs more than two-fold in any cortical layer. Although thalamic inputs are known to evoke about two-fold larger unitary excitatory postsynaptic potentials than cortical ones, the present results suggest that cortical inputs control PV-producing interneurons at least as strongly as thalamic inputs.
Cell type-specific bipolar cell input to ganglion cells in the mouse retina.
Neumann, S; Hüser, L; Ondreka, K; Auler, N; Haverkamp, S
2016-03-01
Many distinct ganglion cell types, which are the output elements of the retina, were found to encode for specific features of a visual scene such as contrast, color information or movement. The detailed composition of retinal circuits leading to this tuning of retinal ganglion cells, however, is apart from some prominent examples, largely unknown. Here we aimed to investigate if ganglion cell types in the mouse retina receive selective input from specific bipolar cell types or if they sample their synaptic input non-selectively from all bipolar cell types stratifying within their dendritic tree. To address this question we took an anatomical approach and immunolabeled retinae of two transgenic mouse lines (GFP-O and JAM-B) with markers for ribbon synapses and type 2 bipolar cells. We morphologically identified all green fluorescent protein (GFP)-expressing ganglion cell types, which co-stratified with type 2 bipolar cells and assessed the total number of bipolar input synapses and the proportion of synapses deriving from type 2 bipolar cells. Only JAM-B ganglion cells received synaptic input preferentially from bipolar cell types other than type 2 bipolar cells whereas the other analyzed ganglion cell types sampled their bipolar input most likely from all bipolar cell terminals within their dendritic arbor.
The UK waste input-output table: Linking waste generation to the UK economy.
Salemdeeb, Ramy; Al-Tabbaa, Abir; Reynolds, Christian
2016-10-01
In order to achieve a circular economy, there must be a greater understanding of the links between economic activity and waste generation. This study introduces the first version of the UK waste input-output table that could be used to quantify both direct and indirect waste arisings across the supply chain. The proposed waste input-output table features 21 industrial sectors and 34 waste types and is for the 2010 time-period. Using the waste input-output table, the study results quantitatively confirm that sectors with a long supply chain (i.e. manufacturing and services sectors) have higher indirect waste generation rates compared with industrial primary sectors (e.g. mining and quarrying) and sectors with a shorter supply chain (e.g. construction). Results also reveal that the construction, mining and quarrying sectors have the highest waste generation rates, 742 and 694 tonne per £1m of final demand, respectively. Owing to the aggregated format of the first version of the waste input-output, the model does not address the relationship between waste generation and recycling activities. Therefore, an updated version of the waste input-output table is expected be developed considering this issue. Consequently, the expanded model would lead to a better understanding of waste and resource flows in the supply chain.
Evaluation of Piloted Inputs for Onboard Frequency Response Estimation
Grauer, Jared A.; Martos, Borja
2013-01-01
Frequency response estimation results are presented using piloted inputs and a real-time estimation method recently developed for multisine inputs. A nonlinear simulation of the F-16 and a Piper Saratoga research aircraft were subjected to different piloted test inputs while the short period stabilator/elevator to pitch rate frequency response was estimated. Results show that the method can produce accurate results using wide-band piloted inputs instead of multisines. A new metric is introduced for evaluating which data points to include in the analysis and recommendations are provided for applying this method with piloted inputs.
Soil-Related Input Parameters for the Biosphere Model
A. J. Smith
2004-09-09
This report presents one of the analyses that support the Environmental Radiation Model for Yucca Mountain Nevada (ERMYN). The ''Biosphere Model Report'' (BSC 2004 [DIRS 169460]) describes the details of the conceptual model as well as the mathematical model and the required input parameters. The biosphere model is one of a series of process models supporting the postclosure Total System Performance Assessment (TSPA) for the Yucca Mountain repository. A schematic representation of the documentation flow for the Biosphere input to TSPA is presented in Figure 1-1. This figure shows the evolutionary relationships among the products (i.e., analysis and model reports) developed for biosphere modeling, and the biosphere abstraction products for TSPA, as identified in the ''Technical Work Plan for Biosphere Modeling and Expert Support'' (TWP) (BSC 2004 [DIRS 169573]). This figure is included to provide an understanding of how this analysis report contributes to biosphere modeling in support of the license application, and is not intended to imply that access to the listed documents is required to understand the contents of this report. This report, ''Soil-Related Input Parameters for the Biosphere Model'', is one of the five analysis reports that develop input parameters for use in the ERMYN model. This report is the source documentation for the six biosphere parameters identified in Table 1-1. The purpose of this analysis was to develop the biosphere model parameters associated with the accumulation and depletion of radionuclides in the soil. These parameters support the calculation of radionuclide concentrations in soil from on-going irrigation or ash deposition and, as a direct consequence, radionuclide concentration in other environmental media that are affected by radionuclide concentrations in soil. The analysis was performed in accordance with the TWP (BSC 2004 [DIRS 169573]) where the governing procedure
Artificial spatiotemporal touch inputs reveal complementary decoding in neocortical neurons.
Oddo, Calogero M; Mazzoni, Alberto; Spanne, Anton; Enander, Jonas M D; Mogensen, Hannes; Bengtsson, Fredrik; Camboni, Domenico; Micera, Silvestro; Jörntell, Henrik
2017-04-04
Investigations of the mechanisms of touch perception and decoding has been hampered by difficulties in achieving invariant patterns of skin sensor activation. To obtain reproducible spatiotemporal patterns of activation of sensory afferents, we used an artificial fingertip equipped with an array of neuromorphic sensors. The artificial fingertip was used to transduce real-world haptic stimuli into spatiotemporal patterns of spikes. These spike patterns were delivered to the skin afferents of the second digit of rats via an array of stimulation electrodes. Combined with low-noise intra- and extracellular recordings from neocortical neurons in vivo, this approach provided a previously inaccessible high resolution analysis of the representation of tactile information in the neocortical neuronal circuitry. The results indicate high information content in individual neurons and reveal multiple novel neuronal tactile coding features such as heterogeneous and complementary spatiotemporal input selectivity also between neighboring neurons. Such neuronal heterogeneity and complementariness can potentially support a very high decoding capacity in a limited population of neurons. Our results also indicate a potential neuroprosthetic approach to communicate with the brain at a very high resolution and provide a potential novel solution for evaluating the degree or state of neurological disease in animal models.
Bayesian robot system identification with input and output noise.
Ting, Jo-Anne; D'Souza, Aaron; Schaal, Stefan
2011-01-01
For complex robots such as humanoids, model-based control is highly beneficial for accurate tracking while keeping negative feedback gains low for compliance. However, in such multi degree-of-freedom lightweight systems, conventional identification of rigid body dynamics models using CAD data and actuator models is inaccurate due to unknown nonlinear robot dynamic effects. An alternative method is data-driven parameter estimation, but significant noise in measured and inferred variables affects it adversely. Moreover, standard estimation procedures may give physically inconsistent results due to unmodeled nonlinearities or insufficiently rich data. This paper addresses these problems, proposing a Bayesian system identification technique for linear or piecewise linear systems. Inspired by Factor Analysis regression, we develop a computationally efficient variational Bayesian regression algorithm that is robust to ill-conditioned data, automatically detects relevant features, and identifies input and output noise. We evaluate our approach on rigid body parameter estimation for various robotic systems, achieving an error of up to three times lower than other state-of-the-art machine learning methods.
Adaptive control for an uncertain robotic manipulator with input saturations
Trong-Toan TRAN; Shuzhi Sam GE; Wei HE
2016-01-01
In this paper, we address the control problem of an uncertain robotic manipulator with input saturations, unknown input scalings and disturbances. For this purpose, a model reference adaptive control like (MRAC-like) is used to handle the input saturations. The model reference is input to state stable (ISS) and driven by the errors between the required control signals and input saturations. The uncertain parameters are dealt with by using linear-in-the-parameters property of robotic dynamics, while unknown input scalings and disturbances are handled by non-regressor based approach. Our design ensures that all the signals in the closed-loop system are bounded, and the tracking error converges to the compact set which depends on the predetermined bounds of the control inputs. Simulation on a planar elbow manipulator with two joints is provided to illustrate the effectiveness of the proposed controller.
Feature-aware natural texture synthesis
Wu, Fuzhang
2014-12-04
This article presents a framework for natural texture synthesis and processing. This framework is motivated by the observation that given examples captured in natural scene, texture synthesis addresses a critical problem, namely, that synthesis quality can be affected adversely if the texture elements in an example display spatially varied patterns, such as perspective distortion, the composition of different sub-textures, and variations in global color pattern as a result of complex illumination. This issue is common in natural textures and is a fundamental challenge for previously developed methods. Thus, we address it from a feature point of view and propose a feature-aware approach to synthesize natural textures. The synthesis process is guided by a feature map that represents the visual characteristics of the input texture. Moreover, we present a novel adaptive initialization algorithm that can effectively avoid the repeat and verbatim copying artifacts. Our approach improves texture synthesis in many images that cannot be handled effectively with traditional technologies.
Micromachined dual input axis rate gyroscope
Juneau, Thor Nelson
The need for inexpensive yet reliable angular rate sensors in fields ranging from automotive to consumer electronics has motivated prolific micromachined rate gyroscope research. The vast majority of research has focused on single input axis rate gyroscopes based upon either translational resonance, such as tuning forks, or structural mode resonance, such as vibrating rings. However, this work presents a novel, contrasting approach based on angular resonance of a rotating rigid rotor suspended by torsional springs. The inherent symmetry of the circular design allows angular rate measurement about two axes simultaneously, hence the name micromachined dual-axis rate gyroscope. The underlying theory of operation, mechanical structure design optimization, electrical interface circuitry, and signal processing are described in detail. Several operational versions were fabricated using two different fully integrated surface micromachining processes as proof of concept. The heart of the dual-axis rate gyroscope is a ˜2 mum thick polysilicon disk or rotor suspended above the substrate by a four beam suspension. When this rotor in driven into angular oscillation about the axis perpendicular to the substrate, a rotation rate about the two axes parallel to the substrate invokes an out of plane rotor tilting motion due to Coriolis acceleration. This tilting motion is capacitively measured and on board integrated signal processing provides two output voltages proportional to angular rate input about the two axes parallel to the substrate. The design process begins with the derivation of gyroscopic dynamics. The equations suggest that tuning sense mode frequencies to the drive oscillation frequency can vastly increase mechanical sensitivity. Hence the supporting four beam suspension is designed such that electrostatic tuning can match modes despite process variations. The electrostatic tuning range is limited only by rotor collapse to the substrate when tuning-voltage induced
Input/output plugin architecture for MDSplus
Stillerman, Joshua, E-mail: jas@psfc.mit.edu [Massachusetts Institute of Technology, 175 Albany Street, Cambridge, MA 02139 (United States); Fredian, Thomas, E-mail: twf@psfc.mit.edu [Massachusetts Institute of Technology, 175 Albany Street, Cambridge, MA 02139 (United States); Manduchi, Gabriele, E-mail: gabriele.manduchi@igi.cnr.it [Consorzio RFX, Euratom-ENEA Association, Corso Stati Uniti 4, Padova 35127 (Italy)
2014-05-15
The first version of MDSplus was released in 1991 for VAX/VMS. Since that time the underlying file formats have remained constant. The software however has evolved, it was ported to unix, linux, Windows, and Macintosh. In 1997 a TCP based protocol, mdsip, was added to provide network access to MDSplus data. In 2011 a mechanism was added to allow protocol plugins to permit the use of other transport mechanisms such as ssh to access data users. This paper describes a similar design which permits the insertion of plugins to handle the reading and writing of MDSplus data at the data storage level. Tree paths become URIs which specify the protocol, host, and protocol specific information. The protocol is provided by a dynamically activated shared library that can provide any consistent subset of the data store access API, treeshr. The existing low level network protocol called mdsip, is activated by defining tree paths like “host::/directory”. Using the new plugin mechanism this is re-implemented as an instance of the general plugin that replaces the low level treeshr input/output routines. It is specified by using a path like “mdsip://host/directory”. This architecture will make it possible to adapt the MDSplus data organization and analysis tools to other underlying data storage. The first new application of this, after the existing network protocol is implemented, will be a plugin based on a key value store. Key value stores, can provide inexpensive scalable, redundant data storage. An example of this might be an Amazon G3 plugin which would let you specify a tree path such as “AG3://container” to access MDSplus data stored in the cloud.
曹留帅; 朱军
2014-01-01
为实现CFD技术在潜艇操纵性优化设计中的应用，文章结合粘性求解器和RBF神经网络预报了潜艇的水动力。通过引入首部和尾部肥瘦指数，确定了潜艇主艇体线型表达的五参数模型。采用均匀试验设计方法，给出了30条潜艇模型的五参数表达。针对每个模型，分别计算了9个漂角下的纵向力、横向力和摇首力矩，得到共计270组数据。为提高计算效率和精度，利用ANSYS ICEM CFD脚本文件和ANSYS FLUENT journal函数实现了从模型建立、网格划分到数值模拟的自动化操作。在多漂角计算过程中，采用“漂角扫掠”方法加快收敛速度。利用上述计算结果训练RBF神经网络，得到了潜艇水动力预报的神经网络模型。以SUBOFF为例，采用该网络预报了其水动力，并与文中数值方法计算结果、试验结果和文献值进行对比，符合较好，说明该方法可应用于工程实践。%To explore the usage of CFD techniques into the optimization design process of submarine maneuverability, CFD-based calculations and RBF neural network were combined to predict the sub-marine hydrodynamics. The fullness of the nose and stern index was introduced to the geometric de-scription of submarine axisymmetric hull, thus creating a five-parameter model for the hull geometry expression. A series of 30 similar hull bodies was adopted by the uniform design approach. For each of the models, 9 different drift angle cases were calculated, and 270 groups of data were achieved con-sisting of the longitudinal force, the lateral force and the yaw moment. To improve the efficiency and accuracy of the computation, automatic mesh and computation using the ANSYS ICEM CFD scripts and ANSYS FLUENT journal functions were used, as well as the drift sweep procedure. A RBF neu-ral network was adopted and trained by the computation results to predict the hydrodynamics of oth-er submarines. For the SUBOFF case, the
Optimal Curiosity-Driven Modular Incremental Slow Feature Analysis.
Kompella, Varun Raj; Luciw, Matthew; Stollenga, Marijn Frederik; Schmidhuber, Juergen
2016-08-01
Consider a self-motivated artificial agent who is exploring a complex environment. Part of the complexity is due to the raw high-dimensional sensory input streams, which the agent needs to make sense of. Such inputs can be compactly encoded through a variety of means; one of these is slow feature analysis (SFA). Slow features encode spatiotemporal regularities, which are information-rich explanatory factors (latent variables) underlying the high-dimensional input streams. In our previous work, we have shown how slow features can be learned incrementally, while the agent explores its world, and modularly, such that different sets of features are learned for different parts of the environment (since a single set of regularities does not explain everything). In what order should the agent explore the different parts of the environment? Following Schmidhuber's theory of artificial curiosity, the agent should always concentrate on the area where it can learn the easiest-to-learn set of features that it has not already learned. We formalize this learning problem and theoretically show that, using our model, called curiosity-driven modular incremental slow feature analysis, the agent on average will learn slow feature representations in order of increasing learning difficulty, under certain mild conditions. We provide experimental results to support the theoretical analysis.
A novel face recognition method with feature combination
LI Wen-shu; ZHOU Chang-le; XU Jia-tuo
2005-01-01
A novel combined personalized feature framework is proposed for face recognition (FR). In the framework, the proposed linear discriminant analysis (LDA) makes use of the null space of the within-class scatter matrix effectively, and Global feature vectors (PCA-transformed) and local feature vectors (Gabor wavelet-transformed) are integrated by complex vectors as input feature of improved LDA. The proposed method is compared to other commonly used FR methods on two face databases (ORL and UMIST). Results demonstrated that the performance of the proposed method is superior to that of traditional FR approaches
Multiclass Bayes error estimation by a feature space sampling technique
Mobasseri, B. G.; Mcgillem, C. D.
1979-01-01
A general Gaussian M-class N-feature classification problem is defined. An algorithm is developed that requires the class statistics as its only input and computes the minimum probability of error through use of a combined analytical and numerical integration over a sequence simplifying transformations of the feature space. The results are compared with those obtained by conventional techniques applied to a 2-class 4-feature discrimination problem with results previously reported and 4-class 4-feature multispectral scanner Landsat data classified by training and testing of the available data.
Multiclass Bayes error estimation by a feature space sampling technique
Mobasseri, B. G.; Mcgillem, C. D.
1979-01-01
A general Gaussian M-class N-feature classification problem is defined. An algorithm is developed that requires the class statistics as its only input and computes the minimum probability of error through use of a combined analytical and numerical integration over a sequence simplifying transformations of the feature space. The results are compared with those obtained by conventional techniques applied to a 2-class 4-feature discrimination problem with results previously reported and 4-class 4-feature multispectral scanner Landsat data classified by training and testing of the available data.
朱懿峰; 宋执环
2011-01-01
An algorithm of RBF neural network soft measurement based on SoC is proposed. The entire training and prediction algorithms are implemented successfully on the hardware platform of dual-core SoC processor. In order to apply algorithm effectively on the platform whose compute speed and memory are limited,specific solutions about the structure of the network, weight update pattern and step and data preprocessing mode is proposed. The data set test results prove that the algorithm transplantation method satisfies the requirements of industrial application,which has a variety of advantages such as portable,low-cost and extensible.%提出了一种面向片上系统(SoC)的RBF神经网络的软测量算法,在OMAP-L137双核处理器SoC 硬件平台上成功实现了整个训练与预测算法.针对SoC计算速度和存储空间等资源有限,对网络结构、权值更新模式和步长以及数据预处理方式等参数提出了具体的解决方案.经过相关数据集的测试结果表明:提出的算法移植方法完全满足工业应用的要求,且具有便携性、低成本、可扩展等多种优点.
何耀耀; 许启发; 杨善林; 余本功
2013-01-01
According to the problem of short-term load forecasting in the power system, this paper proposed a probability density forecasting method using radical basis function (RBF) neural network quantile regression based on the existed researches on combination forecasting and probability interval prediction. The probability density function of load at any period in a day was evaluated. The proposed method can obtain more useful information than point prediction and interval prediction, and can implement the whole probability distribution forecasting for future load. The practical data of a city in China show that the proposed probability density forecasting method can gain more accurate result of point prediction and obtain the forecasting results of integrated probability density function of short-term load.%针对电力系统短期负荷预测问题,在现有的组合预测和概率性区间预测的基础上,提出了基于RBF神经网络分位数回归的概率密度预测方法,得出未来一天中任意时期负荷的概率密度函数,可以得到比点预测和区间预测更多的有用信息,实现了对未来负荷完整概率分布的预测.中国某市实际数据的预测结果表明,提出的概率密度预测方法不仅能得出较为精确的点预测结果,而且能够获得短期负荷完整的概率密度函数预测结果.
张培; 陈光大; 张旭
2011-01-01
It is unnecessary to establish concrete function expression, the known discrete data can be fitted by using neural network to extend hydraulic turbine combined characteristic cure. And we can also add boundary conditions to predict unknown zones, so as to raise the work efficiency and data precision in data treatment concerning hydraulic turbine combined characteristics. This paper intro- duces the use of gP neural network and RBF neural network in extending hydraulic turbine combined characteristic curve. I.astly, the results of the two methods are compared and some conclusions are obtained.%利用神经网络对水轮机综合特性曲线进行数据处理和延伸，不必建立具体的函数关系表达式，就可对已知的离散数据进行拟合。并且还可以结合边界约束条件对未知区域内的数据进行预测，从而提高了水轮机综合特性曲线数据处理的工作效率和数据精度。分别介绍了用BP神经网络和RBF神经网络对水轮机综合特性曲线数据处理和延伸的方法。并采用一机组的样本数据进行训练，比较2种方法的训练结果得出结论。
范立强; 吕国芳
2016-01-01
针对目前混凝土强度预测中存在的不确定性，难以自适应性的确定神经网络隐含层，建立了基于高维云的RBF神经网络的混凝土预测模型。运用MATLAB 8.10进行仿真实验。实验结果表明该模型综合考虑了影响混凝土强度的各种因素，能够实现预测结果的随机性和模糊性，具有更高的预测精度，更快的训练速度，可以广泛应用于生产现场实地的混凝土强度预测和质量检验。%According to the current situation that forecasting concrete strength is uncertainty, it is difficult to make sure the adaptability for the hidden layer of neural network. It has built the prediction model of concrete strength based on high dimensional cloud RBF neural network and has carried out the simulated experiment by MATLAB 8.10. The experimental results show that it considers various factors affecting the concrete strength, can realize randomness and fuzziness of the results, has higher precision of prediction and faster training speed. It can be widely used in the production of in-situ concrete prediction and quality inspection.
李君波; 刘旺开
2014-01-01
针对高压热动力试验台测控系统需要控制的参数多，控制任务要求复杂，控制精度要求高，且各参数间存在耦合的情况，提出了基于 BP＋RBF神经网络 PID的智能控制的方法；应用智能控制的方法解决传统的 PID控制无法解决的问题；实际应用表明基于 BP神经网络整定的 PID控制器具有较好的自学习和自适应性，能保证控制精度等要求，控制效果比较令人满意。%As there are many parameters to be controlled in the Highpressure Thermodynamics Test Unit,the control task is demanding, and the requirement of the control precision is high.And there is a complicated coupling between parameters.In view of these difficulties,a method of intelligent control based on BP + RBF neural network and PID is Proposed.The intelligent control method is used to solve the problem that can not be solved by the traditional PID control method.In practical application the system has met the specifications require-ments perfectly,and achieved excellent controlling effect,though the mathematical model of the control object is complicated.
基于径向基函数网络的宣城市空气质量预测%Based on RBF Neural Network Prediction of Air Quality in the Xuancheng
吴有训; 彭慕平; 刘勇
2011-01-01
A method of prediction of air quality is proposed based on RBF artificial neural network (ANN), making a low-dimension ANN learing matrix through principal component analysis, it could distill the main information of many factorials and remarkably decrease the dimension. According to the characteristics of the prediction of air quality, this paper selects the data in Xuancheng from 2003 to 2005 that greatly influence the air quality. The experimental result shows that the performance of prediction of air quality is favorable, the learning speed is fast and the rate of accurate is high, so it have a practical value. It provides a precise and generalization and efficient way for the prediction of air quality.%提出了一种应用人工神经网络进行空气质量预测的方法,即采用径向基函数神经网络进行短期的空气质量预测；并采用了主成分分析方法降低神经网络学习矩阵维数,浓缩预测信息,降维去噪.选取宣城市气象局2003年到2005年地面气象观测资料作为预测因子,宣城市环境保护监测中心提供的PM10、So2浓度值作为预测对象,进行训练学习和预测验证.研究结果表明:将该方法应用于空气质量预测,效果良好,具有较强的实用性和推广能力.
Rapid Airplane Parametric Input Design (RAPID)
Smith, Robert E.
1995-01-01
RAPID is a methodology and software system to define a class of airplane configurations and directly evaluate surface grids, volume grids, and grid sensitivity on and about the configurations. A distinguishing characteristic which separates RAPID from other airplane surface modellers is that the output grids and grid sensitivity are directly applicable in CFD analysis. A small set of design parameters and grid control parameters govern the process which is incorporated into interactive software for 'real time' visual analysis and into batch software for the application of optimization technology. The computed surface grids and volume grids are suitable for a wide range of Computational Fluid Dynamics (CFD) simulation. The general airplane configuration has wing, fuselage, horizontal tail, and vertical tail components. The double-delta wing and tail components are manifested by solving a fourth order partial differential equation (PDE) subject to Dirichlet and Neumann boundary conditions. The design parameters are incorporated into the boundary conditions and therefore govern the shapes of the surfaces. The PDE solution yields a smooth transition between boundaries. Surface grids suitable for CFD calculation are created by establishing an H-type topology about the configuration and incorporating grid spacing functions in the PDE equation for the lifting components and the fuselage definition equations. User specified grid parameters govern the location and degree of grid concentration. A two-block volume grid about a configuration is calculated using the Control Point Form (CPF) technique. The interactive software, which runs on Silicon Graphics IRIS workstations, allows design parameters to be continuously varied and the resulting surface grid to be observed in real time. The batch software computes both the surface and volume grids and also computes the sensitivity of the output grid with respect to the input design parameters by applying the precompiler tool
Remote Sensing Image Feature Extracting Based Multiple Ant Colonies Cooperation
Zhang Zhi-long
2014-02-01
Full Text Available This paper presents a novel feature extraction method for remote sensing imagery based on the cooperation of multiple ant colonies. First, multiresolution expression of the input remote sensing imagery is created, and two different ant colonies are spread on different resolution images. The ant colony in the low-resolution image uses phase congruency as the inspiration information, whereas that in the high-resolution image uses gradient magnitude. The two ant colonies cooperate to detect features in the image by sharing the same pheromone matrix. Finally, the image features are extracted on the basis of the pheromone matrix threshold. Because a substantial amount of information in the input image is used as inspiration information of the ant colonies, the proposed method shows higher intelligence and acquires more complete and meaningful image features than those of other simple edge detectors.
Holmes, Jon L.
1999-05-01
The Features area of JCE Online is now readily accessible through a single click from our home page. In the Features area each column is linked to its own home page. These column home pages also have links to them from the online Journal Table of Contents pages or from any article published as part of that feature column. Using these links you can easily find abstracts of additional articles that are related by topic. Of course, JCE Online+ subscribers are then just one click away from the entire article. Finding related articles is easy because each feature column "site" contains links to the online abstracts of all the articles that have appeared in the column. In addition, you can find the mission statement for the column and the email link to the column editor that I mentioned above. At the discretion of its editor, a feature column site may contain additional resources. As an example, the Chemical Information Instructor column edited by Arleen Somerville will have a periodically updated bibliography of resources for teaching and using chemical information. Due to the increase in the number of these resources available on the WWW, it only makes sense to publish this information online so that you can get to these resources with a simple click of the mouse. We expect that there will soon be additional information and resources at several other feature column sites. Following in the footsteps of the Chemical Information Instructor, up-to-date bibliographies and links to related online resources can be made available. We hope to extend the online component of our feature columns with moderated online discussion forums. If you have a suggestion for an online resource you would like to see included, let the feature editor or JCE Online (jceonline@chem.wisc.edu) know about it. JCE Internet Features JCE Internet also has several feature columns: Chemical Education Resource Shelf, Conceptual Questions and Challenge Problems, Equipment Buyers Guide, Hal's Picks, Mathcad
Echterhoff, J.; Simonis, I.; Atkinson, R.
2012-04-01
The infrastructure to gather, store and access information about our environment is improving and growing rapidly. The increasing amount of information allows us to get a better understanding of the current state of our environment, historical processes and to simulate and predict the future state of the environment. Finer grained spatial and temporal data and more reliable communications make it easier to model dynamic states and ephemeral features. The exchange of information within and across geospatial domains is facilitated through the use of harmonized information models. The Observations & Measurements (O&M) developed through OGC and standardised by ISO is an example of such a cross-domain information model. It is used in many domains, including meteorology, hydrology as well as the emergency management. O&M enables harmonized representation of common metadata that belong to the act of determining the state of a feature property, whether by sensors, simulations or humans. In addition to the resulting feature property value, information such as the result quality but especially the time that the result applies to the feature property can be represented. Temporal metadata is critical to modelling past and future states of a feature. The features, and the semantics of each property, are defined in domain specific Application Schema using the General Feature Model (GFM) from ISO 19109 and usually encoded following ISO 19136. However, at the moment these standards provide only limited support for the representation and handling of time varying feature data. Features like rivers, wildfires or gas plumes have a defined state - for example geographic extent - at any given point in time. To keep track of changes, a more complex model for example using time-series coverages is required. Furthermore, the representation and management of feature property value changes via the service interfaces defined by OGC and ISO - namely: WFS and WCS - would be rather complex
Reduction of Feature Vectors Using Rough Set Theory for Human Face Recognition
Bhattacharjee, Debotosh; Nasipuri, Mita; Kundu, M
2010-01-01
In this paper we describe a procedure to reduce the size of the input feature vector. A complex pattern recognition problem like face recognition involves huge dimension of input feature vector. To reduce that dimension here we have used eigenspace projection (also called as Principal Component Analysis), which is basically transformation of space. To reduce further we have applied feature selection method to select indispensable features, which will remain in the final feature vectors. Features those are not selected are removed from the final feature vector considering them as redundant or superfluous. For selection of features we have used the concept of reduct and core from rough set theory. This method has shown very good performance. It is worth to mention that in some cases the recognition rate increases with the decrease in the feature vector dimension.
Uncertainty of input data for room acoustic simulations
Jeong, Cheol-Ho; Marbjerg, Gerd; Brunskog, Jonas
2016-01-01
summarizes potential advanced absorption measurement techniques that can improve the quality of input data for room acoustic simulations. Lastly, plenty of uncertain input data are copied from unreliable sources. Software developers and users should be careful when spreading such uncertain input data. More......Although many room acoustic simulation models have been well established, simulation results will never be accurate with inaccurate and uncertain input data. This study addresses inappropriateness and uncertainty of input data for room acoustic simulations. Firstly, the random incidence absorption......-included input data are proven to produce perceptually noticeable changes in the objective parameters, such as the sound pressure level, and loudness-based reverberation time. Surfaces should not be assumed to be locally reacting, particularly for multi-layered absorbers having air cavities. Secondly...
Cui, Yunfeng; Bai, Jing
2005-01-01
Liver kinetic study of [18F]2-fluoro-2-deoxy-D-glucose (FDG) metabolism in human body is an important tool for functional modeling and glucose metabolic rate estimation. In general, the arterial blood time-activity curve (TAC) and the tissue TAC are required as the input and output functions for the kinetic model. For liver study, however, the arterial-input may be not consistent with the actual model input because the liver has a dual blood supply from the hepatic artery (HA) and the portal vein (PV) to the liver. In this study, the result of model parameter estimation using dual-input function is compared with that using arterial-input function. First, a dynamic positron emission tomography (PET) experiment is performed after injection of FDG into the human body. The TACs of aortic blood, PV blood, and five regions of interest (ROIs) in liver are obtained from the PET image. Then, the dual-input curve is generated by calculating weighted sum of both the arterial and PV input curves. Finally, the five liver ROIs' kinetic parameters are estimated with arterial-input and dual-input functions respectively. The results indicate that the two methods provide different parameter estimations and the dual-input function may lead to more accurate parameter estimation.
Analytical delay models for RLC interconnects under ramp input
REN Yinglei; MAO Junfa; LI Xiaochun
2007-01-01
Analytical delay models for Resistance Inductance Capacitance (RLC)interconnects with ramp input are presented for difierent situations,which include overdamped,underdamped and critical response cases.The errors of delay estimation using the analytical models proposed in this paper are less bv 3%in comparison to the SPICE-computed delay.These models are meaningful for the delay analysis of actual circuits in which the input signal is ramp but not ideal step input.
The Application of Input Theory to English Classroom Teaching
刘坤
2015-01-01
Early in the 1980s, Stephen Krashen has proposed a comprehensive and overall Input Theory that explains how the sec⁃ond language is acquired. It is still very referential to present English classroom teaching. In this essay, applications of Input Theo⁃ry to English classroom teaching are developed from six aspects, involving the nature of second language acquisition, comprehensi⁃ble input and so on.
Knowledge Management in Customer Integration: A Customer Input Management System
Füller, Kathrin; Abud, Elias; Böhm, Markus; Krcmar, Helmut
2016-01-01
Customers can take an active role in the innovation process and provide their input (e.g., ideas, idea evaluations, or complaints) to the different phases of the innovation process. However, the management of a huge amount of unstructured customer input poses a challenge for companies. Existing software solutions focus on the early stages of idea management, and neglect the interoperability of tools, sharing, and reuse of customer inputs across innovation cycles and departments. Following the...
Connelly, Elizabeth B.; Allen, Craig R.; Hatfield, Kirk; Palma-Oliveira, José M.; Woods, David D.; Linkov, Igor
2017-02-20
The National Academy of Sciences (NAS) definition of resilience is used here to organize common concepts and synthesize a set of key features of resilience that can be used across diverse application domains. The features in common include critical functions (services), thresholds, cross-scale (both space and time) interactions, and memory and adaptive management. We propose a framework for linking these features to the planning, absorbing, recovering, and adapting phases identified in the NAS definition. The proposed delineation of resilience can be important in understanding and communicating resilience concepts.
杨剑锋; 张翠; 张峰
2015-01-01
针对机械臂运动轨迹控制中存在的跟踪精度不高的问题，采用了一种基于EC-RBF神经网络的模型参考自适应控制方案对机械臂进行模型辨识与轨迹跟踪控制。该方案采用了两个RBF神经网络，运用EC-RBF学习算法，采用离线与在线相结合的方法来训练神经网络，一个用来实现对机械臂进行模型辨识，一个用来实现对机械臂轨迹跟踪控制。对二自由度机械臂进行仿真，结果表明，使用该控制方案对机械臂进行轨迹跟踪控制具有较高的控制精度，且因采用EC-RBF学习算法使网络具有更快的训练速度，从而使得控制过程较迅速。%According to the problem that the tracking accuracy is not high enough in trajectory tracking control of robot manipulators, a model reference adaptive control scheme based on EC-RBF neural networks is adopted to achieve robot manipulator model identification and trajectory tracking control. This control scheme contains two RBF neural networks which are trained offline and online, using EC-RBF learning algorithm. The one is used to identify the robot manipulator’s model, and the other one is used to achieve its trajectory tracking control. Simulation result of 2-degree-of-freedom robot manipulator demonstrates that using this method for robot manipulator trajectory tracking control has high control accuracy, and the networks which gain high training speed because of the EC-RBF learning algorithm make the control process faster.
黄江涛; 高正红; 白俊强; 周铸; 赵轲
2014-01-01
The vetex deformation pattern of multi-block grids is developed based on RBF technique cou-pling with Delaunay graphic mapping to establishe grid deformation platform.The Delaunay cell is generated using the grid sparse points and block vetexs.When aerodynamic configuration deforms,the block vetex will be moved by RBF technique and the Delaunay cells will be deformed,consequently,the mapping relation could be obtained to realize grid deformation.A certain lateral aerobus is taken as an typical example,jig shape is proposed firstly,static aeroelastic analysis is processed then to get the cruise shape of jig configura-tion,and the comparison with cruise design shape is investigated.The results show that the correction of jig shape design method established in this paper is validated.On the other hand,the “mapping surface”can be used to realized CFD/CSD data transformation effectively.%基于 RBF 径向基函数建立了多块对接网格块顶点运动模式，进一步耦合 Delaunay 图映射建立网格变形技术；利用基准网格的稀疏序列节点以及区域顶点建立 Delaunay 四面体映射单元，利用 RBF 技术操作区域顶点运动，使映射单元进行变形，从而应用映射关系实现网格变形；针对某型支线客机进行了型架外形设计，并进一步进行对型架外形进行静气动弹性计算与设计巡航外形对比，验证了所建立的型架外形设计方法的正确性；静气动弹性计算中，通过建立共用的“映射曲面”，实现 CFD/CSD 数据高效交换。
Krashen’s Input Hypothesis and Foreign Language Teaching
彭辉
2013-01-01
Krashen’s Input Hypothesis is one of the most important theories in second language acquisition.The theory provides a good theoretical framework for foreign language teaching in China.The paper introduces the basic ideas of Krashen’s second language acquisition theories,the concept of comprehensible input,and Krashen’s interpretation of input hypothesis.Thus,this paper aims to study Krashen’s Comprehensible Input and attempts to discover how to facilitate China’s foreign language teaching.
Electronically Tunable High Input Impedance Voltage-Mode Multifunction Filter
Chen, Hua-Pin; Yang, Wan-Shing
A novel electronically tunable high input impedance voltage-mode multifunction filter with single inputs and three outputs employing two single-output-operational transconductance amplifiers, one differential difference current conveyor and two capacitors is proposed. The presented filter can be realized the highpass, bandpass and lowpass functions, simultaneously. The input of the filter exhibits high input impedance so that the synthesized filter can be cascaded without additional buffers. The circuit needs no any external resistors and employs two grounded capacitors, which is suitable for integrated circuit implementation.
The effects of redundant control inputs in optimal control
DUAN ZhiSheng; HUANG Lin; YANG Ying
2009-01-01
For a stabillzable system,the extension of the control inputs has no use for stabllizability,but it is important for optimal control.In this paper,a necessary and sufficient condition is presented to strictly decrease the quadratic optimal performance index after control input extensions.A similar result is also provided for H_2 optimal control problem.These results show an essential difference between single-input and multi-input control systems.Several examples are taken to illustrate related problems.
Characteristic of energy input for laser forming sheet metal
Liqun Li(李俐群); Yanbin Chen(陈彦宾); Xiaosong Feng(封小松)
2003-01-01
Laser forming is a process in which laser-induced thermal deformation is used to form sheet metal withouta hard forming tool or external forces. The energy input of laser beam is the key factor for the temperatureand stress distribution of sheet metal. The purpose of this work is to investigate the influence of energyinput condition on heat input and deformation angle for two-dimension laser forming. Variations in heatinput resulting from material deformation was calculated and discussed in this paper at first. Furthermore,in laser forming under the condition of constant laser energy input, the effects of energy input mode ondeformation angle and temperature field were investigated.
Effect of correlated inputs on DO (dissolved oxygen) uncertainty
Brown, L.C.; Song, Q.
1988-06-01
Although uncertainty analysis has been discussed in recent water-quality-modeling literature, much of the work has assumed that all input variables and parameters are mutually independent. The objective of this paper is to evaluate the importance of correlation among the model inputs in the study of model-output uncertainty. The model used for demonstrating the influence of input-variable correlation is the Streeter-Phelps dissolved oxygen equation. The model forms the basis of many of the water-quality models currently in use and the relationships between model inputs and output-state variables are well understood.
Orthogonal topography in the parallel input architecture of songbird HVC.
Elliott, Kevin C; Wu, Wei; Bertram, Richard; Hyson, Richard L; Johnson, Frank
2017-06-15
Neural activity within the cortical premotor nucleus HVC (acronym is name) encodes the learned songs of adult male zebra finches (Taeniopygia guttata). HVC activity is driven and/or modulated by a group of five afferent nuclei (the Medial Magnocellular nucleus of the Anterior Nidopallium, MMAN; Nucleus Interface, NIf; nucleus Avalanche, Av; the Robust nucleus of the Arcopallium, RA; the Uvaeform nucleus, Uva). While earlier evidence suggested that HVC receives a uniformly distributed and nontopographic pattern of afferent input, recent evidence suggests this view is incorrect (Basista et al., ). Here, we used a double-labeling strategy (varying both the distance between and the axial orientation of dual tracer injections into HVC) to reveal a massively parallel and in some cases topographic pattern of afferent input. Afferent neurons target only one rostral or caudal location within medial or lateral HVC, and each HVC location receives convergent input from each afferent nucleus in parallel. Quantifying the distributions of single-labeled cells revealed an orthogonal topography in the organization of afferent input from MMAN and NIf, two cortical nuclei necessary for song learning. MMAN input is organized across the lateral-medial axis whereas NIf input is organized across the rostral-caudal axis. To the extent that HVC activity is influenced by afferent input during the learning, perception, or production of song, functional models of HVC activity may need revision to account for the parallel input architecture of HVC, along with the orthogonal input topography of MMAN and NIf. © 2017 Wiley Periodicals, Inc.
Nguyen-Duy, Khiem; Petersen, Lars Press; Knott, Arnold;
2014-01-01
This paper presents the design of a 300-Watt isolated power supply for MOS gate driver circuit in medium and high voltage applications. The key feature of the developed power supply is having a very low circuit input-to-output parasitic capacitance, thus maximizing its noise immunity. This makes ...
Environmental Transport Input Parameters for the Biosphere Model
M. A. Wasiolek
2003-06-27
This analysis report is one of the technical reports documenting the Environmental Radiation Model for Yucca Mountain Nevada (ERMYN), a biosphere model supporting the total system performance assessment (TSPA) for the geologic repository at Yucca Mountain. A graphical representation of the documentation hierarchy for the ERMYN is presented in Figure 1-1. This figure shows relationships among the reports developed for biosphere modeling and biosphere abstraction products for the TSPA, as identified in the ''Technical Work Plan: for Biosphere Modeling and Expert Support'' (TWP) (BSC 2003 [163602]). Some documents in Figure 1-1 may be under development and not available when this report is issued. This figure provides an understanding of how this report contributes to biosphere modeling in support of the license application (LA), but access to the listed documents is not required to understand the contents of this report. This report is one of the reports that develops input parameter values for the biosphere model. The ''Biosphere Model Report'' (BSC 2003 [160699]) describes the conceptual model, the mathematical model, and the input parameters. The purpose of this analysis is to develop biosphere model parameter values related to radionuclide transport and accumulation in the environment. These parameters support calculations of radionuclide concentrations in the environmental media (e.g., soil, crops, animal products, and air) resulting from a given radionuclide concentration at the source of contamination (i.e., either in groundwater or volcanic ash). The analysis was performed in accordance with the TWP (BSC 2003 [163602]). This analysis develops values of parameters associated with many features, events, and processes (FEPs) applicable to the reference biosphere (DTN: M00303SEPFEPS2.000 [162452]), which are addressed in the biosphere model (BSC 2003 [160699]). The treatment of these FEPs is described in BSC (2003 [160699
Spatiotemporal Features for Asynchronous Event-based Data
Xavier eLagorce
2015-02-01
Full Text Available Bio-inspired asynchronous event-based vision sensors are currently introducing a paradigm shift in visual information processing. These new sensors rely on a stimulus-driven principle of light acquisition similar to biological retinas. They are event-driven and fully asynchronous, thereby reducing redundancy and encoding exact times of input signal changes, leading to a very precise temporal resolution. Approaches for higher-level computer vision often rely on the realiable detection of features in visual frames, but similar definitions of features for the novel dynamic and event-based visual input representation of silicon retinas have so far been lacking. This article addresses the problem of learning and recognizing features for event-based vision sensors, which capture properties of truly spatiotemporal volumes of sparse visual event information. A novel computational architecture for learning and encoding spatiotemporal features is introduced based on a set of predictive recurrent reservoir networks, competing via winner-take-all selection. Features are learned in an unsupervised manner from real-world input recorded with event-based vision sensors. It is shown that the networks in the architecture learn distinct and task-specific dynamic visual features, and can predict their trajectories over time.
Coherent branching feature bisimulation
Tessa Belder
2015-04-01
Full Text Available Progress in the behavioral analysis of software product lines at the family level benefits from further development of the underlying semantical theory. Here, we propose a behavioral equivalence for feature transition systems (FTS generalizing branching bisimulation for labeled transition systems (LTS. We prove that branching feature bisimulation for an FTS of a family of products coincides with branching bisimulation for the LTS projection of each the individual products. For a restricted notion of coherent branching feature bisimulation we furthermore present a minimization algorithm and show its correctness. Although the minimization problem for coherent branching feature bisimulation is shown to be intractable, application of the algorithm in the setting of a small case study results in a significant speed-up of model checking of behavioral properties.
Coherent branching feature bisimulation
T. Belder (Tessa); M.H. ter Beek (Maurice); E.P. de Vink (Erik Peter)
2015-01-01
textabstractProgress in the behavioral analysis of software product lines at the family level benefits from further development of the underlying semantical theory. Here, we propose a behavioral equivalence for feature transition systems (FTS) generalizing branching bisimulation for labeled
Fingerprint Feature Extraction Algorithm
Mehala. G
2014-03-01
Full Text Available The goal of this paper is to design an efficient Fingerprint Feature Extraction (FFE algorithm to extract the fingerprint features for Automatic Fingerprint Identification Systems (AFIS. FFE algorithm, consists of two major subdivisions, Fingerprint image preprocessing, Fingerprint image postprocessing. A few of the challenges presented in an earlier are, consequently addressed, in this paper. The proposed algorithm is able to enhance the fingerprint image and also extracting true minutiae.
Fingerprint Feature Extraction Algorithm
Mehala. G
2014-01-01
The goal of this paper is to design an efficient Fingerprint Feature Extraction (FFE) algorithm to extract the fingerprint features for Automatic Fingerprint Identification Systems (AFIS). FFE algorithm, consists of two major subdivisions, Fingerprint image preprocessing, Fingerprint image postprocessing. A few of the challenges presented in an earlier are, consequently addressed, in this paper. The proposed algorithm is able to enhance the fingerprint image and also extractin...
BEDOPS: high-performance genomic feature operations.
Neph, Shane; Kuehn, M Scott; Reynolds, Alex P; Haugen, Eric; Thurman, Robert E; Johnson, Audra K; Rynes, Eric; Maurano, Matthew T; Vierstra, Jeff; Thomas, Sean; Sandstrom, Richard; Humbert, Richard; Stamatoyannopoulos, John A
2012-07-15
The large and growing number of genome-wide datasets highlights the need for high-performance feature analysis and data comparison methods, in addition to efficient data storage and retrieval techniques. We introduce BEDOPS, a software suite for common genomic analysis tasks which offers improved flexibility, scalability and execution time characteristics over previously published packages. The suite includes a utility to compress large inputs into a lossless format that can provide greater space savings and faster data extractions than alternatives. http://code.google.com/p/bedops/ includes binaries, source and documentation.
Driver Fatigue Features Extraction
Gengtian Niu
2014-01-01
Full Text Available Driver fatigue is the main cause of traffic accidents. How to extract the effective features of fatigue is important for recognition accuracy and traffic safety. To solve the problem, this paper proposes a new method of driver fatigue features extraction based on the facial image sequence. In this method, first, each facial image in the sequence is divided into nonoverlapping blocks of the same size, and Gabor wavelets are employed to extract multiscale and multiorientation features. Then the mean value and standard deviation of each block’s features are calculated, respectively. Considering the facial performance of human fatigue is a dynamic process that developed over time, each block’s features are analyzed in the sequence. Finally, Adaboost algorithm is applied to select the most discriminating fatigue features. The proposed method was tested on a self-built database which includes a wide range of human subjects of different genders, poses, and illuminations in real-life fatigue conditions. Experimental results show the effectiveness of the proposed method.
Chen, Zhe; Rau, Pei-Luen Patrick
2017-03-01
This study presented two experiments on Chinese handwriting performance (time, accuracy, the number of protruding strokes and number of rewritings) and subjective ratings (mental workload, satisfaction, and preference) on mobile devices. Experiment 1 evaluated the effects of size of the input box, input method and display size on Chinese handwriting performance and preference. It was indicated that the optimal input sizes were 30.8 × 30.8 mm, 46.6 × 46.6 mm, 58.9 × 58.9 mm and 84.6 × 84.6 mm for devices with 3.5-inch, 5.5-inch, 7.0-inch and 9.7-inch display sizes, respectively. Experiment 2 proved the significant effects of location of the input box, input method and display size on Chinese handwriting performance and subjective ratings. It was suggested that the optimal location was central regardless of display size and input method.
Pan, Weixing X; Mao, Tianyi; Dudman, Joshua T
2010-01-01
The basal ganglia play a critical role in the regulation of voluntary action in vertebrates. Our understanding of the function of the basal ganglia relies heavily upon anatomical information, but continued progress will require an understanding of the specific functional roles played by diverse cell types and their connectivity. An increasing number of mouse lines allow extensive identification, characterization, and manipulation of specified cell types in the basal ganglia. Despite the promise of genetically modified mice for elucidating the functional roles of diverse cell types, there is relatively little anatomical data obtained directly in the mouse. Here we have characterized the retrograde labeling obtained from a series of tracer injections throughout the dorsal striatum of adult mice. We found systematic variations in input along both the medial-lateral and anterior-posterior neuraxes in close agreement with canonical features of basal ganglia anatomy in the rat. In addition to the canonical features we have provided experimental support for the importance of non-canonical inputs to the striatum from the raphe nuclei and the amygdala. To look for organization at a finer scale we have analyzed the correlation structure of labeling intensity across our entire dataset. Using this analysis we found substantial local heterogeneity within the large-scale order. From this analysis we conclude that individual striatal sites receive varied combinations of cortical and thalamic input from multiple functional areas, consistent with some earlier studies in the rat that have suggested the presence of a combinatorial map.
Weixing X Pan
2010-12-01
Full Text Available The basal ganglia play a critical role in the regulation of voluntary action in vertebrates. Our understanding of the function of the basal ganglia relies heavily upon anatomical information, but continued progress will require an understanding of the specific functional roles played by diverse cell types and their connectivity. An increasing number of mouse lines allow extensive identification, characterization, and, manipulation of specified cell types in the basal ganglia. Despite the promise of genetically modified mice for elucidating the functional roles of diverse cell types, there is relatively little anatomical data obtained directly in the mouse. Here we have characterized the retrograde labeling obtained from a series of tracer injections throughout the dorsal striatum of adult mice. We found systematic variations in input along both the medial-lateral and anterior-posterior neuraxes in close agreement with canonical features of basal ganglia anatomy in the rat. In addition to the canonical features we have provided experimental support for the importance of non-canonical inputs to the striatum from the raphe nuclei and the amygdala. To look for organization at a finer scale we have analyzed the correlation structure of labeling intensity across our entire dataset. Using this analysis we found substantial local heterogeneity within the large-scale order. From this analysis we conclude that individual striatal sites receive varied combinations of cortical and thalamic input from multiple functional areas, consistent with some earlier studies in the rat that have suggested the presence of a combinatorial map.
Mundiñano, Iñaki-Carril; Hernandez, Maria; Dicaudo, Carla; Ordoñez, Cristina; Marcilla, Irene; Tuñon, Maria-Teresa; Luquin, Maria-Rosario
2013-09-01
Olfactory impairment is a common feature of neurodegenerative diseases such as Parkinson's disease (PD), Alzheimer's disease (AD) and dementia with Lewy bodies (DLB). Olfactory bulb (OB) pathology in these diseases shows an increased number of olfactory dopaminergic cells, protein aggregates and dysfunction of neurotransmitter systems. Since cholinergic denervation might be a common underlying pathophysiological feature, the objective of this study was to determine cholinergic innervation of the OB in 27 patients with histological diagnosis of PD (n = 5), AD (n = 14), DLB (n = 8) and 8 healthy control subjects. Cholinergic centrifugal inputs to the OB were clearly reduced in all patients, the most significant decrease being in the DLB group. We also studied cholinergic innervation of the OB in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated monkeys (n = 7) and 7 intact animals. In MPTP-monkeys, we found that cholinergic innervation of the OB was reduced compared to control animals (n = 7). Interestingly, in MPTP-monkeys, we also detected a loss of cholinergic neurons and decreased dopaminergic innervation in the horizontal limb of the diagonal band, which is the origin of the centrifugal cholinergic input to the OB. All these data suggest that cholinergic damage in the OB might contribute, at least in part, to the olfactory dysfunction usually exhibited by these patients. Moreover, decreased cholinergic input to the OB found in MPTP-monkeys suggests that dopamine depletion in itself might reduce the cholinergic tone of basal forebrain cholinergic neurons.
Feature detection techniques for preprocessing proteomic data.
Sellers, Kimberly F; Miecznikowski, Jeffrey C
2010-01-01
Numerous gel-based and nongel-based technologies are used to detect protein changes potentially associated with disease. The raw data, however, are abundant with technical and structural complexities, making statistical analysis a difficult task. Low-level analysis issues (including normalization, background correction, gel and/or spectral alignment, feature detection, and image registration) are substantial problems that need to be addressed, because any large-level data analyses are contingent on appropriate and statistically sound low-level procedures. Feature detection approaches are particularly interesting due to the increased computational speed associated with subsequent calculations. Such summary data corresponding to image features provide a significant reduction in overall data size and structure while retaining key information. In this paper, we focus on recent advances in feature detection as a tool for preprocessing proteomic data. This work highlights existing and newly developed feature detection algorithms for proteomic datasets, particularly relating to time-of-flight mass spectrometry, and two-dimensional gel electrophoresis. Note, however, that the associated data structures (i.e., spectral data, and images containing spots) used as input for these methods are obtained via all gel-based and nongel-based methods discussed in this manuscript, and thus the discussed methods are likewise applicable.
What to measure next to improve decision making? On top-down task driven feature saliency
Hansen, Lars Kai; Karadogan, Seliz; Marchegiani, Letizia
2011-01-01
Top-down attention is modeled as decision making based on incomplete information. We consider decisions made in a sequential measurement situation where initially only an incomplete input feature vector is available, however, where we are given the possibility to acquire additional input values a...
Flexible Peripheral Component Interconnect Input/Output Card
Bigelow, Kirk K.; Jerry, Albert L.; Baricio, Alisha G.; Cummings, Jon K.
2010-01-01
The Flexible Peripheral Component Interconnect (PCI) Input/Output (I/O) Card is an innovative circuit board that provides functionality to interface between a variety of devices. It supports user-defined interrupts for interface synchronization, tracks system faults and failures, and includes checksum and parity evaluation of interface data. The card supports up to 16 channels of high-speed, half-duplex, low-voltage digital signaling (LVDS) serial data, and can interface combinations of serial and parallel devices. Placement of a processor within the field programmable gate array (FPGA) controls an embedded application with links to host memory over its PCI bus. The FPGA also provides protocol stacking and quick digital signal processor (DSP) functions to improve host performance. Hardware timers, counters, state machines, and other glue logic support interface communications. The Flexible PCI I/O Card provides an interface for a variety of dissimilar computer systems, featuring direct memory access functionality. The card has the following attributes: 8/16/32-bit, 33-MHz PCI r2.2 compliance, Configurable for universal 3.3V/5V interface slots, PCI interface based on PLX Technology's PCI9056 ASIC, General-use 512K 16 SDRAM memory, General-use 1M 16 Flash memory, FPGA with 3K to 56K logical cells with embedded 27K to 198K bits RAM, I/O interface: 32-channel LVDS differential transceivers configured in eight, 4-bit banks; signaling rates to 200 MHz per channel, Common SCSI-3, 68-pin interface connector.
TANG Zhipeng; GONG Peiping; LIU Weidong; LI Jiangsu
2015-01-01
Industrial wastewater discharge in China is increasing with the country's economic development and it is worthy of concern.The discharge is primarily relevant to the direct discharge coefficient of each sector of the economy,its direct input coefficient and the final demand in input-output models.In this study,we calculated the sensitivity of the reduction in the Chinese industrial wastewater discharge using the direct input coefficients based on the theory of error-transmission in an input-outpnt framework.Using input-output models,we calculated the direct and total industrial wastewater discharge coefficients.Analysis of 2007 input-output data of 30 sectors of the Chinese economy and of 30 provincial regions of China indicates that by lowering their direct input coefficients,the manufacturers of textiles,paper and paper products,chemical products,smelting and metal pressing,telecommunication equipment,computers and other electronic equipment will significantly reduce their amounts of industrial wastewater discharge.By lowering intra-provincial direct input coefficients to industrial sectors themselves of Jiangsu,Shandong and Zhejiang,there will be a significant reduction in industrial wastewater discharge for the country as a whole.Investment in production technology and improvement in organizational efficiency in these sectors and in these provinces can help lessen the direct input coefficients,thereby effectively achieving a reduction in industrial wastewater discharge in China via industrial restructuring.
Feature Technology in Product Modeling
ZHANG Xu; NING Ruxin
2006-01-01
A unified feature definition is proposed. Feature is form-concentrated, and can be used to model product functionalities, assembly relations, and part geometries. The feature model is given and a feature classification is introduced including functional, assembly, structural, and manufacturing features. A prototype modeling system is developed in Pro/ENGINEER that can define the assembly and user-defined form features.
Input--output capital coefficients for energy technologies. [Input-output model
Tessmer, R.G. Jr.
1976-12-01
Input-output capital coefficients are presented for five electric and seven non-electric energy technologies. They describe the durable goods and structures purchases (at a 110 sector level of detail) that are necessary to expand productive capacity in each of twelve energy source sectors. Coefficients are defined in terms of 1967 dollar purchases per 10/sup 6/ Btu of output from new capacity, and original data sources include Battelle Memorial Institute, the Harvard Economic Research Project, The Mitre Corp., and Bechtel Corp. The twelve energy sectors are coal, crude oil and gas, shale oil, methane from coal, solvent refined coal, refined oil products, pipeline gas, coal combined-cycle electric, fossil electric, LWR electric, HTGR electric, and hydroelectric.
Phaff, H. [TNO Bouw en Ondergrond, Delft (Netherlands)
2010-04-15
Buildings in reality, use more energy than predicted. Among many causes, occupant behaviour plays an important role. Better simulation of occupant behaviour, with respect to thermal comfort and energy use of buildings, opens the possibility to design better, comfortable buildings that have lower energy consumption in reality. Thermal discomfort, a dynamical version of Fanger's PPD, is proposed to be used as input to simulate occupant behaviour via a 'flexible task list' and two Markov processes. [Dutch] Simulatie van bewonersgedrag m.b.t. energiegebruik in gebouwen biedt de mogelijkheid om gebouwen en bijbehorende energie installaties zo te ontwerpen dat ze prettiger zijn om in te wonen en te werken. Thermisch discomfort, een dynamische versie van PPD (percentage of dissatisfied persons) wordt hier voorgesteld om via een Markov-proces en een takenlijst bewonersgedrag mee te simuleren.
Fast increases in urban sewage inputs to rivers of Indonesia
Suwarno, D.; Löhr, A.; Kroeze, C.; Widianarko, B.
2014-01-01
We present estimates for nitrogen (N) and phosphorus (P) sewage inputs to 19 Indonesian rivers for 1970–2050. Future trends are based on the four scenarios of the Millennium Ecosystem Assessment. Our results indicate a rapid increase in N and P pollution from sewage over time. In 1970, N and P input