Energy Technology Data Exchange (ETDEWEB)

An artificial neural network (ANN), trained as an inverse of the controlled plant, to function as a power system stabilizer (PSS) is presented in this paper. In order to make the proposed ANN PSS work properly, it was trained over the full working range of the generating unit with a large variety of disturbances. Data used to train the ANN PSS consists of the control input and the synchronous machine response with an adaptive PSS (APSS) controlling the generator. During training, the ANN was required to memorize the reverse input/output mapping of the synchronous machine. After the training, the output of the synchronous machine was applied as the input of the ANN PSS and the output of the ANN PSS was used as the control signal. Simulation results show that the proposed ANN PSS can provide good damping of the power system over a wide ...

International Nuclear Information System (INIS)

Critical power characteristics of tight lattice rod assembly was investigated using a simple-shaped experimental apparatus. An electrically heated rod with four spacers was placed in a circular tube, and boiling transition condition for a rod in an annular geometry was clarified varing annulus clearance. It was found that critical heat flux depends strongly on the clearance accoding as the gap becomes smaller. This results was compared with KfK correlation and the trends were well correlated. (author).

Energy Technology Data Exchange (ETDEWEB)

A novel way to extract relevant parameters associated with the outgoing ions from nuclear reactions, obtained by digitizing the signals provided by a Bragg curve spectrometer (BCS) is presented. This allowed the implementation of a more thorough pulse-shape analysis. Due to the complexity of this task, it was required to take advantage of new and more powerful computational paradigms. This was fulfilled using a back-propagation artificial neural network (ANN) as a pattern identifier. Over training of ANNs is a common problem during the training stage. In the performance of the ANN there is a compromise between its size and the size of the training set. Here, this effect will be illustrated in relation to the problem of Bragg Curve (BC) identification. (Author)

Energy Technology Data Exchange (ETDEWEB)

This paper presents a systematic approach for designing a self-tuning power system stabilizer (PSS) based on artificial neural network (ANN). An ANN is used for self-tuning the parameters of PSS in real-time. The nodes in the input layer of the ANN receive generator terminal active power (P), reactive power (Q), and voltage (V{sub t}), while the nodes in the output layer provide the optimum PSS parameters, e.g. stabilizing gain (K{sub STAB}), time constants (T{sub 1} and T{sub 2}). A new approach for the selection of number of neurons in the hidden layer has been proposed. Investigations reveal that the dynamic performance of the system with self-tuning PSS based on ANN (ST-ANNPSS) is quite robust over a wide range of loading conditions and equivalent reactance, X{sub e}. (Author)

International Nuclear Information System (INIS)

The aim of the study was the attempt to evaluate the influence of two different methods of cardiac perfusion SPECT reconstruction (FBP and ITW) on clinical efficacy in diagnosing the coronary artery disease as well as the cardiac ischemia detection in three areas of heart vascularized by main coronary arteries: LAD, LCX and RCA with the use of artificial neural networks (ANN). The study was performed retrospectively with the use of the diagnostic image records as well as clinical dataset of 43 patients. Myocardial perfusion stress/rest SPECT study and X-ray coronarography data were evaluated for each patient. The results of coronary angiography were considered the reference method. The cardiac SPECT data were reconstructed using the two different methods: filtered backprojection (FBP) and iterative Wallis method (ITW). The local perfusion deficits denominated in stress and rest study in three main vessel cardiac segments were the main input values for the ...

Energy Technology Data Exchange (ETDEWEB)

The aim of this study is to develop the methodology which enables to identify the mechanical properties of element such as stress intensity factor by using the AE parameters. Considering the multivariate and nonlinear properties of AE parameters such as ringdown count, rise time, energy, event duration and peak amplitude from fatigue cracks of machine element the principal component regression(PCR) and artificial neural network(ANN) models for the estimation of stress intensity factor were developed and validated. The AE parameters were found to be very significant to estimate the stress intensity factor. Since the statistical values including correlation coefficients, standard mr of calibration, standard error of prediction and bias were stable, the PCR and ANN models for stress intensity factor were very robust. The performance of ANN model for unknown data of stress intensity factor was better than that of PCR model

International Nuclear Information System (INIS)

Precise measurements of the single spin asymmetry AN, and the double spin asymmetry ANN, in proton-proton (pp) elastic scattering in the region of four-momentum transfer squared 0.0012 have been performed using a polarized atomic hydrogen gas jet target and the Relativistic Heavy Ion Collider (RHIC) polarized proton beam. We present measurements of AN and ANN at center-of-mass energies ?(s)=6.8 and 13.7 GeV. These spin-dependent observables are sensitive to the poorly known hadronic spin-dependent amplitudes. Comparing AN at different energies, a ?(s) dependence of the hadronic single spin-flip amplitude is suggested. A hadronic double spin-flip amplitude from the ANN data is consistent with zero within a 2-? level. We also present ??T, estimated from the measured ANN data. The results for ??T are consistent with zero. Our results provide significant constraints toward a comprehensive understanding of ...

Energy Technology Data Exchange (ETDEWEB)

Relationships of ultimate and proximate analysis of 4540 US coal samples from 25 states with gross calorific value (GCV) have been investigated by regression and artificial neural networks (ANNs) methods. Three set of inputs: (a) volatile matter, ash and moisture (b) C, H, N, O, S and ash (c) C, H{sub exclusive} {sub of} {sub moisture}, N, O{sub exclusive} {sub of} {sub moisture}, S, moisture and ash were used for the prediction of GCV by regression and ANNs. The multivariable regression studies have shown that the model (c) is the most suitable estimator of GCV. Running of the best arranged ANNs structures for the models (a) to (c) and assessment of errors have shown that the ANNs are not better or much different from regression, as a common and understood technique, in the prediction of uncomplicated relationships between proximate and ultimate analysis and coal GCV. (author)

International Nuclear Information System (INIS)

On the problem of alarm when parts are falling in nuclear power plant, the artificial neural network (ANN) alarm method based on the signal time-frequency characteristics was developed. The method was realized by the improved BP algorithm, and demonstrated with the data from simulation experiments

Energy Technology Data Exchange (ETDEWEB)

Artificial Neural Networks (ANNs) are parallel distributed processing machines. The unique characteristics of ANNs are: Fault tolerance, robustness, plasticity and generalization. These offer great potential in many AI applications such as character recognition. Handwritten character recognition is an intrinsically interesting problem, but the difficulties of this task are the many variations in the characters. A robust new incremental learning method, which combines supervised and unsupervised learning paradigms implemented by the Functional Link Net, is illustrated with experimental results. Clustering, based on unsupervised learning, classifies the input data into several categories. The supervised learning paradigm then further classifies the data in the clustered categories.

Energy Technology Data Exchange (ETDEWEB)

The use of electricity is indispensable to modern life. As Macao Special Administrative Region becomes a gaming and tourism center in Asia, modeling the consumption of electricity is critical to Macao's economic development. The purposes of this paper are to conduct an extensive literature review on modeling of electricity consumption, and to identify key climatic, demographic, economic and/or industrial factors that may affect the electricity consumption of a country/city. It was identified that the five factors, namely temperature, population, the number of tourists, hotel room occupancy and days per month, could be used to characterize Macao's monthly electricity consumption. Three selected approaches including multiple regression, artificial neural network (ANN) and wavelet ANN were used to derive mathematical models of the electricity consumption. The accuracy of these models was assessed by using the mean squared error ...

Energy Technology Data Exchange (ETDEWEB)

The paper presents an alternative approach for the studies of high voltage transmission lines based on artificial intelligence and more specifically artificial neural networks (ANNs). In contrast to the existing conventional-analytical techniques and simulations which are using in the calculations empirical and/or approximating equations, this approach is based only on actual field data and actual measurements. The proposed approach is applied on high voltage transmission lines in order to calculate the lightning outages, on grounding systems in order to assess the grounding resistance and on high voltage transmission lines' polluted insulators in order to estimate the critical flashover voltage. The obtained results are very close to the actual ones for all three case studies, something which clearly implies that the ANN approach is well working and has an acceptable accuracy, constituting an additional tool of electric engineers. ...

Energy Technology Data Exchange (ETDEWEB)

Field investigations of major and damaging ice jamming on the Saint John River at Dickey, Maine , and at Sainte-Anne de-Madawaska, New Brunswick, in 1991, were described. The investigations included measurement of water surface profiles and shear wall heights at both sites. The measurements were supplemented by information from local observers and data collection agencies. Using a simplified equilibrium analysis, ice jam thickness and water level at the Dickey site was found to be generally in agreement with observed values. At Sainte-Anne-de-Madawaska sufficient data was obtained to construct and calibrate the numerical model RIVJAM which determined the configuration of the jam in nonequilibrium reaches. Use of the model enabled the successful reproduction of a measured water profile along the jam and the prediction of the approximate thickness of the jam, which was generally less than the measured shear wall height. 14 refs., 11 figs.

Energy Technology Data Exchange (ETDEWEB)

Distance protection, differential protection and directional comparison schemes are presently used for protecting transmission lines. Directional comparison relays are set to respond to faults in the protection zone without intentional time delay and are, therefore, used where high-speed fault clearing is needed. Artificial Neural Networks (ANNs) can handle most situations which cannot be defined sufficiently for finding a deterministic solution. The design and testing of an ANN for directional comparison protection of transmission lines are presented in this paper. Training patterns were generated using voltage and current samples for faults at various locations along a transmission line. The faults were simulated using an electromagnetic transient program and a sample three-phase power system. The performance of the proposed discriminator was checked using data simulated for testing and the fault data recorded from 240 kV and 500 kV lines. ...

International Nuclear Information System (INIS)

Growing of PV for electricity generation is one of the highest in the field of the renewable energies and this tendency is expected to continue in the next years. Due to the various seasonal, hourly and daily changes in climate, it is relatively difficult to find a suitable analytic model for predicting the performance of a grid-connected photovoltaic (GCPV) plant. In this paper, an artificial neural network is used for modelling and predicting the power produced by a 20 kWp GCPV plant installed on the roof top of the municipality of Trieste (latitude 45 deg. 40'N, longitude 13 deg. 46'E), Italy. An experimental database of climate (irradiance and air temperature) and electrical (power delivered to the grid) data from January 29th to May 25th 2009 has been used. Two ANN models have been developed and implemented on experimental climate and electrical data. The first one is a multivariate model based on the solar irradiance and the air temperature, while the second ...

British Library Electronic Table of Contents (United Kingdom)

The objective of this study is to develop an artificial neural network (ANN) model to predict the thermal conductivity of ethylene glycol-water solutions based on experimentally measured variables. The thermal conductivity of solutions at different concentrations and various temperatures was measured using the cylindrical cell method that physical properties of the solution are being determined fills the annular space between two concentric cylinders. During the experiment, heat flows in the radial direction outwards through the test liquid filled in the annual gap to cooling water. In the steady state, conduction inside the cell was described by the Fourier equation in cylindrical coordinates, with boundary conditions corresponding to heat transfer between the solution and cooling water. ...

British Library Electronic Table of Contents (United Kingdom)

A marine propulsion system is a very complicated system composed of many mechanical components. As a result, the vibration signal of a gearbox in the system is strongly coupled with the vibration signatures of other components including a diesel engine and main shaft. It is therefore imperative to assess the coupling effect on diagnostic reliability in the process of gear fault diagnosis. For this reason, a fault detection and diagnosis method based on bispectrum analysis and artificial neural networks (ANNs) was proposed for the gearbox with consideration given to the impact of the other components in marine propulsion systems. To monitor the gear conditions, the bispectrum analysis was first employed to detect gear faults. The amplitude-frequency plots containing gear characteristic sign...

CERN Document Server

The Elman artificial neural network (ANN) (feedback connection) was used for seismic data filtering. The recurrent connection that characterizes this network offers the advantage of storing values from the previous time step to be used in the current time step. The proposed structure has the advantage of training simplicity by a back-propagation algorithm (steepest descent). Several trials were addressed on synthetic (with 10% and 50% of random and Gaussian noise) and real seismic data using respectively 10 to 30 neurons and a minimum of 60 neurons in the hidden layer. Both an iteration number up to 4000 and arrest criteria were used to obtain satisfactory performances. Application of such networks on real data shows that the filtered seismic section was efficient. Adequate cross-validation test is done to ensure the performance of network on new data sets.

Energy Technology Data Exchange (ETDEWEB)

A power system stabilizer based on GMV (Generalized Minimum Variance), one of the adaptive control techniques, is developed to enhance the dynamic performances of a power system using an Artificial Neural Network (ANN). The stabilizer consists of two parts. One part is Inverse Dynamics Neural Networks (IDNN), which is trained to identify the inverse dynamics of controlled plant and used as a one-step ahead controller, or inverse controller. The other part is Adaptive Reference Model (ARM), which prevents excessive controller output. The ARM produces the modified reference value by minimizing a cost function recursively on the assumption that the IDNN perfectly identifies the controlled plant. The IDNN is used in the minimization procedure to calculate the sensitivities. The proposed controller is simulated in a typical one-machine-infinite-bus power system to show its effectiveness to damp sustained low frequency oscillation. (author)

CERN Document Server

Evolutionary artificial neural networks (EANNs) refer to a special class of artificial neural networks (ANNs) in which evolution is another fundamental form of adaptation in addition to learning. Evolutionary algorithms are used to adapt the connection weights, network architecture and learning algorithms according to the problem environment. Even though evolutionary algorithms are well known as efficient global search algorithms, very often they miss the best local solutions in the complex solution space. In this paper, we propose a hybrid meta-heuristic learning approach combining evolutionary learning and local search methods (using 1st and 2nd order error information) to improve the learning and faster convergence obtained using a direct evolutionary approach. The proposed technique is tested on three different chaotic time series and the test results are compared with some popular neuro-fuzzy systems and a recently developed cutting angle method of global ...

Energy Technology Data Exchange (ETDEWEB)

With using artificial neural networks (ANNs), an analytical study related to the heated length effect on critical heat flux (CHF) has been carried out to make an improvement of the CHF prediction accuracy based on local condition correlations or table. It has been carried out to suggest a feasible criterion of the threshold length-to-diameter (L/D) value in which heated length could affect CHF. And within the criterion, a L/D correction factor has been developed through conventional regression. In order to validate the developed L/D correction factor, CHF experiments for various heated lengths have been carried out under low and intermediate pressure conditions. The developed threshold L/D correlation provides a new feasible criterion of L/D threshold value. The developed correction factor gives a reasonable accuracy for the original database, showing the error of -2.18% for average and 27.75% for RMS, and promising results for new experimental data. 7 refs., 12 ...

International Nuclear Information System (INIS)

Research highlights: #-># The building occupancy affecting the cooling load prediction is studied. #-># PENN model is adopted in this study for predicting the building cooling load. #-># Statistical approach is adopted to result a less prejudice prediction performance. #-># Results show that occupancy data can significantly improve the prediction. -- Abstract: Building cooling load prediction is one of the key factors in the success of energy-saving measures. Many computational models available in the industry today have been developed from either forward or inverse modeling approaches. However, most of these models require extensive computer resources and involve lengthy computation. This paper discusses the use of data-driven intelligent approaches, a probabilistic entropy-based neural (PENN) model to predict the cooling load of a building. Although it is common knowledge that the presence and activity of building occupants have a significant impact on the required ...