By Topic

Neural Networks to Power Systems, 1993. ANNPS '93., Proceedings of the Second International Forum on Applications of

Date 1993

Filter Results

Displaying Results 1 - 25 of 77
  • Short-term load forecasting by artificial neural networks using individual and collective data of preceding years

    Publication Year: 1993 , Page(s): 245 - 250
    Cited by:  Papers (5)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (530 KB)  

    This paper presents a short-term load forecasting technique for summer using an artificial neural network (ANN). The purpose of this study is to forecast accurately daily peak load for a target period using actual data from the same period of the previous several years as training data. This paper describes two methods. In one method, the actual data of each year for the several years earlier are used for each ANN. The other method uses the collective data of several years for the training of the ANN. With the proposed method, the mean absolute forecasting error was below 2%. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Short-term system load forecasting using an artificial neural network

    Publication Year: 1993 , Page(s): 239 - 244
    Cited by:  Papers (8)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (715 KB)  

    This paper presents a new, artificial neural network (ANN) based model for the calculation of next day's load forecasts. The model's most significant aspects fall into the following two areas: training process and selection of the input variables. Insights gained during the development of the model regarding the choice of the input variables, and their transformations, the design of the ANN structure, the selection of the training cases and the training process itself are described in the paper. The new model has been tested under a wide variety of conditions and it is shown in this paper to produce excellent results. Comparison results between an existing regression-based model that is currently in production use and the ANN model are very encouraging. The ANN model consistently outperforms the existing model in terms of both average errors over a long period of time and number of 'large' errors. Conclusions reached from this development are sufficiently general to be used by other electric power utilities. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Short-term load forecasting using an artificial neural network

    Publication Year: 1993 , Page(s): 233 - 238
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (519 KB)  

    This paper discusses an artificial neural network (ANN) model for short-term load forecasting. A two-step training method to cope with a shortage of training data and overfitting problems is proposed. A limit is conducted to the range where the ANN's weights are allowed to change in order to preserve the general relation between the inputs and the output of the ANN. The ANN trained with this two-step training method demonstrates improved accuracy over conventional methods, including ANNs which employ ordinary training algorithms. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Short-term load forecasting using diagonal recurrent neural network

    Publication Year: 1993 , Page(s): 227 - 232
    Cited by:  Papers (7)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (502 KB)  

    This paper presents a new approach for short term load forecasting using a diagonal recurrent neural network with an adaptive learning rate. The fully connected recurrent neural network (FRNN), where all neurons are coupled to one another, is difficult to train and to converge in a short time. The DRNN is a modified model of FRNN. It requires fewer weights than FRNN and rapid convergence has been demonstrated. A dynamic backpropagation algorithm coupled with an adaptive learning rate guarantees even faster convergence. To consider the effect of seasonal load variation on the accuracy of the proposed forecasting model, forecasting accuracy is evaluated throughout a whole year. Simulation results show that the forecast accuracy is improved. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An adaptive fuzzy logic controller for AC-DC power systems

    Publication Year: 1993 , Page(s): 218 - 223
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (410 KB)  

    The paper presents a new approach to the design of a supplementary stabilizing controller for a HVDC transmission link using fuzzy logic. The fuzzy controller relates significant and observable variables like speed and its rate of the generator speed and its rate of change of the generator to a control signal for the rectifier current regulator loop using fuzzy membership functions. These variables evaluate the control rules using the compositional rules of inference. The fuzzy controller is equivalent to a nonlinear PI controller, whose gains are adapted depending on the error and its rate of change. The effectiveness of the proposed controller is demonstrated by simulation studies on a DC transmission link connected to a weak AC power system and subjected to transient disturbances. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Experimental studies on micro-computer based fuzzy logic power system stabilizer

    Publication Year: 1993 , Page(s): 212 - 217
    Cited by:  Papers (12)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (433 KB)  

    A microcomputer based fuzzy logic power system stabilizer is implemented to an actual hydroelectric generator with the rating of 5.25 MVA to investigate its efficiency in real time control. The stabilizing signal is determined by using sampled real power signals to damp the system oscillations. The results show the proposed stabilizer improves the system damping effectively subject to various types of disturbances. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neuro-fuzzy controller for enhancing the performance of extinction angle control of inverters in a MTDC-AC system

    Publication Year: 1993 , Page(s): 206 - 211
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (438 KB)  

    Constant extinction angle control of an inverter in a MTDC-AC system is of utmost importance for proper operation under all contingencies. In this paper, the process of control is treated as a pattern recognition problem. A neuro-fuzzy controller is implemented and used for online operation of a MTDC-AC system to enhance the performance of extinction angle control. The proposed controller has significantly improved the system performance for cases studied. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On fuzzy control based static VAr compensator for power system stability control

    Publication Year: 1993 , Page(s): 201 - 205
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (346 KB)  

    The paper presents an application of fuzzy control to determine the control signal of a static VAr compensator (SVC) for improving power system stability. The quantity of reactive power that should be supplied/absorbed by the SVC is calculated depending on the error and the change of error of the electrical power output at each sampling time. The control signal is calculated using fuzzy membership functions. The effectiveness of the proposed control method is demonstrated by a one machine infinite bus system. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Discrimination of partial discharge from noise in XLPE cable lines using a neural network

    Publication Year: 1993 , Page(s): 193 - 198
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (414 KB)  

    This paper describes an experimental study of the discrimination of partial discharge (PD) signals from external noise in a cross-linked polyethylene (XLPE) power cable by using a neural network (NN) system. Measurement of PD signal and external noise was carried out with a PD pulse recorder for a 66 kV XLPE cable with an artificial defect and a drill. The NN was a three-layer artificial neural system with feedforward connections, and its learning method was a backpropagation algorithm. Its input information was a combination of the discharge magnitude, the number of pulse counts, and the phase angle of applied voltage. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fuzzy logic based automatic diagnosis of power apparatus by infrared imaging

    Publication Year: 1993 , Page(s): 187 - 192
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (509 KB)  

    This paper describes a thermographic application system for the electrical power industry. The infrared imager has a range from 40 degrees C to 950 degrees C and maximum resolution down to 0.01 degrees C. A new algorithm for image matching has been devised to match slightly different infrared images of the same object by adaptively adjusting the five parameters, namely x- and y- translation, rotation, x- and y- scaling respectively. The diagnosis is automatically executed by a fuzzy logic-based expert system which extracts the major features within the thermograms and recommends appropriate actions for maintenance. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Abnormality diagnosis of GIS using adaptive resonance theory

    Publication Year: 1993 , Page(s): 181 - 186
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (563 KB)  

    The paper presents an artificial neural network (ANN) approach using ART2 (Adaptive Resonance Theory 2) to a diagnostic system for gas insulated switchgear (GIS). To begin with, the authors show the background of abnormality diagnosis of GISs from the view point of predictive maintenance of them. Then, they discuss the necessity of ART-type ANNs, as an unsupervised learning method, in which neuron(s) are self-organized and self-created when detecting unexpected signals even if untrained by ANNs through a sensor. Finally, they present brief simulation results and their evaluation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Probabilistic diagnosis of power system nodal voltages with ART2

    Publication Year: 1993 , Page(s): 177 - 180
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (356 KB)  

    This paper proposes a new method for nodal voltage diagnosis in power systems using a self-organization artificial neural network. ART2 is utilized to classify power system conditions. A probability voltage security index is evaluated by the resulting classification. The proposed method is used for tracking the voltage profile continuously. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Structural control based on genetic algorithm and neural network for electric power systems

    Publication Year: 1993 , Page(s): 169 - 174
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (466 KB)  

    This paper presents a method of structural control of electric power networks for improving their stability. The method is based on the FACTS concept, a genetic algorithm and neural network. FACTS equipment will provide some new ways for improving stability by controlling the reactance of transmission lines in terms of structure control of the power network. A case study with a multimachine power system is presented and discussed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal VAr allocation by genetic algorithm

    Publication Year: 1993 , Page(s): 163 - 168
    Cited by:  Papers (7)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (614 KB)  

    Keeping up with the times and computer technology, many researchers have applied new mathematical approaches extensively to solve various problems in power systems. AI technology, fuzzy theory and artificial neural networks are recent trends. This paper presents a new optimization method for reactive power planning using genetic algorithms. The genetic algorithm (GA) is a kind of search algorithm based on the mechanics of natural selection and genetics. This algorithm can search for a global solution using a multiple path and have a structure fit to integer problems. The proposed method was applied to practical 51-bus and 224-bus systems to show its feasibility and capabilities. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal economic power dispatch using genetic algorithms

    Publication Year: 1993 , Page(s): 157 - 162
    Cited by:  Papers (10)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (651 KB)  

    This paper presents the genetic algorithm approach to adaptive optimal economic dispatch of electrical power systems. Genetic algorithms, also termed as the machine learning approach to artificial intelligence, are powerful stochastic optimization techniques with potential features of random search, hill climbing, statistical sampling and competition. Genetic algorithmic approach to power system optimization, as reported here for a case of economic power dispatch, consists essentially of minimizing the objective function while gradually satisfying the constraint relations. The unique problem solving strategy of the genetic algorithm and their suitability for power system optimization is described. The advantages of the genetic algorithmic approach in terms of problem reduction, flexibility and solution methodology are also discussed. The suitability of the proposed approach is described for the case of a 15 generator power system. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Distribution systems copper and iron loss minimization by genetic algorithm

    Publication Year: 1993 , Page(s): 151 - 156
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (532 KB)  

    This paper proposes a new GA method to minimize distribution system losses including power transformer iron loss. Since the transformer iron loss is approximately proportional to the square of a transformer's primary voltage, one can minimize the sum of transformer iron loss and line resistive loss by adjusting line voltages and line currents appropriately. Since the problem is formulated as a complex combinatorial optimization problem, it is solved by applying a genetic algorithm (GA) in this paper. Several numerical examples are shown to demonstrate the proposed method. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A genetic algorithm based approach to economic load dispatching

    Publication Year: 1993 , Page(s): 145 - 150
    Cited by:  Papers (7)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (411 KB)  

    This paper presents a two-phase genetic algorithm for economic load dispatching of generators in power systems. The problem of ELD is expressed as a Lagrange function. The conventional GA has a drawback that the algorithm is not so effective as the number of variables increases. To improve the GA characteristic, a two-phase GA is proposed to obtain better solutions. The proposed genetic algorithm may be applied to minimize the Lagrange function with respect to the generator unit output. The effectiveness of the proposed method is demonstrated in a 20-unit system. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural network based power system transient stability criterion using DSP-PC system

    Publication Year: 1993 , Page(s): 136 - 141
    Cited by:  Papers (2)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (333 KB)  

    Transient stability assessment plays an important role in power systems. The transient stability deals with the electromechanical oscillation of synchronous generators, created by a disturbance in the power system. For example, in the case of a transmission line fault, assume that faulted line section is first isolated and then reclosed (reclosure); there then exists a threshold parameter known as the stable critical clearing time (CCT). This paper describes a neural network based adaptive pattern recognition approach for estimation of the critical clearing time. Numerical examples are presented to illustrate this approach. In the neural network considered in this research work, a multi DSP-PC system (digital signal processor-personal computer system) is used for realizing faster backpropagation by applying pipeline operation and parallel operation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Transient stability evaluation using an artificial neural network (power systems)

    Publication Year: 1993 , Page(s): 130 - 135
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (554 KB)  

    This paper describes a power system transient stability evaluation method using an artificial neural network (ANN). To improve the accuracy of the evaluation, the authors propose a new type of training signal which is a reciprocal of the action time of a step-out relay (SOR) after the fault occurrence. In simulation results of a 16-bus system, the evaluation accuracy of the ANN trained using the proposed training signal is about 20 percent more accurate than that of an ANN trained using the conventional 0/1 digital signal. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning tangent hypersurfaces for fast assessment of transient stability

    Publication Year: 1993 , Page(s): 124 - 129
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (490 KB)  

    A new direct method for transient security assessment of multimachine power systems is presented. A local approximation of the stability boundary is made by tangent hypersurfaces which are developed from Taylor series expansion of the transient energy function in the state space nearby a certain class of unstable equilibrium points (UEP). Two approaches for an estimation of the stability region are proposed by taking into account the second order coefficients or alternatively, the second and third order coefficients of the hypersurfaces. Results for two representative power systems are described and a comparison is made with the hyperplane method, demonstrating the superiority of the proposed approach and its potential in real power system applications. Artificial neural networks are used to determine the unknown coefficients of the hypersurfaces independently of operating conditions. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Piecewise linear factor analysis by four layer neural networks and its application for modeling the partial discharge data

    Publication Year: 1993 , Page(s): 475 - 480
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (404 KB)  

    This paper presents the methodology of a nonlinear version of factor analysis by four layer feedforward neural networks and, as an example of its application, the result of modeling the structure of partial discharge data measured on a power cable. Here, the authors introduce the four layer auto associative memory with a reduced size of its second layer that learns identity mapping (the same pattern is used for both of the input data and the supervisory data for the network) and is used for data compression of the multivariate data, then they show that it is valid as a tool for so-called 'piecewise linear factor analysis'. They demonstrate the advantages of the piecewise linear factor analysis method over the current linear scheme regarding the modeling of the unknown structure of multivariate data such as electric pulse distribution data generated by simulated partial discharge. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A systematic search method for obtaining multiple local optimal solutions of nonlinear programming problems

    Publication Year: 1993 , Page(s): 467 - 474
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (688 KB)  

    The authors propose a systematic method to find several local minima for general nonlinear optimizatioin problems. They develop some analytical results for a quasi-gradient system and reflected gradient system and apply them to explore the topological aspects of the critical points of the objective function. By properly switching between a quasi-gradient system and a reflected gradient system, the proposed method can obtain a set of local minima. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning algorithm for neural networks by solving nonlinear equations

    Publication Year: 1993 , Page(s): 461 - 466
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (541 KB)  

    The BP (backpropagation) process is a popular learning algorithm for neural networks. Despite of many successful applications, the BP process has some known drawbacks. These drawbacks stem from that the BP process is a gradient based optimization procedure without a linear search. In this paper, a new learning algorithm is presented based on a solution method of nonlinear equations. Compared with the former optimization procedure, the proposed method often converges faster to desired results. Newton's method is basically applied to solve the nonlinear equations. However, the major difficulty with Newton's method is that its convergence depends on an initial point. In order to assure a global convergence, independent of an initial point, the Homotopy continuation method is employed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Application of artificial neural networks in adaptive interlocking systems

    Publication Year: 1993 , Page(s): 453 - 458
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (522 KB)  

    Interlocks have been in use ever since protective relaying schemes were implemented for power devices like generators, transformers, transmission lines, etc. Although the science of protective relaying has undergone marked changes and improvements, the interlocking philosophy has not changed much. Recently with the availability of programmable logic controllers (PLCs), interlocking schemes have been implemented by means of these devices with basic philosophy of logic remaining the same. This paper suggests the implementation of interlocking schemes with artificial neural networks employing threshold logic unit (TLU) elements. It is demonstrated that while the basic hardware required is same as that of any common PLC, the suggested system will have added flexibility, adaptability to various switchyard modifications, electrical topology changes and equipment/switchyard conditions as well as network complexity. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive relaying using artificial neural network

    Publication Year: 1993 , Page(s): 447 - 452
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (471 KB)  

    Adaptive relaying has the validity for a wide variety of applications. Here a typical problem of maloperation is considered. The application of the modified multilayer perceptron (MLP) mode can successfully avoid the maloperation of a relay. For the cases considered, it shows encouraging results. The advantage associated with the presented MLP model is that the modified characteristic can be defined in the absence of a definite analytical model since the artificial neural network (ANN) can learn it through input-output patterns. The methodology can be extended to many adaptive protective schemes. This report just opens new vistas for the exploration of the application of ANNs in adaptive protective schemes, and further investigations could lead to increased confidence. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.