Scheduled System Maintenance:
Some services will be unavailable Sunday, March 29th through Monday, March 30th. We apologize for the inconvenience.
By Topic

Neural Networks to Power Systems, 1991., Proceedings of the First International Forum on Applications of

Date 23-26 July 1991

Filter Results

Displaying Results 1 - 25 of 57
  • Proceedings of the First International Forum on Applications of Neural Networks to Power Systems (Cat. No.91TH0374-9)

    Publication Year: 1991
    Save to Project icon | Request Permissions | PDF file iconPDF (13 KB)  
    Freely Available from IEEE
  • A solution of generation expansion problem by means of neutral network

    Publication Year: 1991 , Page(s): 219 - 224
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (392 KB)  

    The authors present how to solve power system generation expansion planning by artificial neutral networks, especially the Hopfield type network. In the first place, generation expansion planning is formulated as a 0-1 integer programming problem and then mapped onto the modified Hopfield neural network that can handle a large number of inequality constraints. The neural network simulated on a digital computer can solve a fairly large problem of 20 units over 10 periods. Although the network cannot give the optimal solution, the results obtained are quite promising View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Query based learning in a multilayered perceptron in the presence of data jitter

    Publication Year: 1991 , Page(s): 72 - 75
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (204 KB)  

    Stochastically perturbed feature data is said to be jittered. Jittered data has a convolutional smoothing effect in the classification (or regression) space. Parametric knowledge of the jitter can be used to perturb the training cost function of the neural network so that more efficient training can be performed. The improvement is more striking when the addended cost function is used in a query based learning procedure View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A hybrid neural network and expert system for monitoring fossil fuel power plants

    Publication Year: 1991 , Page(s): 215 - 218
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (196 KB)  

    A fully recurrent neural network and a rule-based expert system are combined in a hybrid architecture to provide power plant operators with an intelligent on-line advisory system. Its purpose is to alert the operator to impending or occurring abnormal conditions related to the plant's boiler. The hybrid system is trained to provide a model of the boiler under normal operation, while the rules address a general set of diagnostic events. Deviation from normal conditions trigger rules to suggest corrective action. This system is intended to increase plant availability and efficiency by automatically deducing abnormal boiler conditions before they become critical View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Hybrid expert system neural network hierarchical architecture for classifying power system contingencies

    Publication Year: 1991 , Page(s): 76 - 82
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (596 KB)  

    The authors present a hierarchical architecture which couples an expert system (ES) with multiple neural networks (NNs) for classifying power system contingencies. The ES performs the `coarse' screening to decide if a contingency is potentially harmful and then determines its type of security limit violations. It uses a set of heuristic rules and a set of performance indicators to filter out the secure contingencies and direct the potentially harmful ones for further analysis in the appropriate NN. The NN's take the coarse classification outcome from the ES and perform a `finer' screening by classifying the contingencies according to the severity of limit violations View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A perspective on use of neural-net computing in training simulator design

    Publication Year: 1991 , Page(s): 210 - 214
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (404 KB)  

    The authors explore and demonstrate the feasibility of combined artificial intelligence/neural-net methodology for carrying out dynamic power system analysis in real-time. This methodology will be capable of characterizing the near term transient stability of the system, as well as perform mid-term and long term dynamic security analyses. In the transient stability analysis, the authors are principally concerned with a question whether the system can return to the steady state. In the mid-term and long-term-security analysis, they are also concerned with a manner in which the final steady state is reached, whether system performance constraints are violated on the way and whether further protective actions might be triggered unexpectedly with undesired actions View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Genetic algorithms approach to voltage optimization

    Publication Year: 1991 , Page(s): 139 - 143
    Cited by:  Papers (11)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (356 KB)  

    The authors consider the use of genetic algorithms as a measure of voltage optimization of electric power system. Genetic algorithms are optimization and learning techniques based on natural selection and natural population genetics. A formation of a power system is encoded to a string of characters called an artificial chromosome the initial population of strings are generated at random, and then they are evolved by a genetic algorithm. The experiments with the prototype implementation are presented. These results verified the feasibility of genetic algorithms approach to power engineering View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On-line training of neural network model and controller for turbogenerators

    Publication Year: 1991 , Page(s): 161 - 165
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (288 KB)  

    The authors are concerned with the development of a neural network (NN) regulator for turbogenerator adaptive control. The NN regulator is designed based on a hierarchical architecture of neural networks. The back-propagation (BP) algorithm is used hierarchically in the NN regulator for on-line training of the turbogenerator NN model and controller. Dynamic modelling of the turbogenerator system has been investigated using the multilayer NN. The NN regulator has been implemented on a simulated complex nonlinear turbogenerator system. Simulation results evaluating the performance of the NN regulator under different operation conditions and disturbances are presented 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 to unit commitment

    Publication Year: 1991 , Page(s): 256 - 260
    Cited by:  Papers (6)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (348 KB)  

    Artificial neural networks are currently being applied to a variety of complex combinatorial optimization and nonlinear programming problems. In this paper, a combination of Hopfield Tank type, and Chua-Lin type artificial neural networks is applied to solve simultaneously the unit commitment and the associated economic unit dispatch problems. The approach is based on imbedding the various constraints in a generalized energy function, and then defining the network dynamics in such a way that the generalized energy function is a Lyapunov function of the artificial neural network. The novel feature of the proposed approach is that the nonlinear programming and the combinatorial optimization problems are solved simultaneously by one network. An illustrative example is also presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An artificial neural-net based method for estimating power system dynamic stability index

    Publication Year: 1991 , Page(s): 129 - 133
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (340 KB)  

    The author presents an artificial neural net based method for evaluating power system dynamic stability. An adaptive pattern recognition technique is utilized to estimate an index for power system dynamic stability so that computational efforts are reduced and numerical instability problems are avoided. The proposed method is based on a multi-layer feedforward perceptron View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A neural networks approach to voltage security monitoring and control

    Publication Year: 1991 , Page(s): 89 - 93
    Cited by:  Papers (2)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (416 KB)  

    Voltage collapse evaluation methods require elaborate computations to determine the existence of feasible load flow solutions in power systems. The time-consuming process of solving the stiff nonlinear system equations in these evaluation methods makes them inefficient for on-line monitoring of voltage collapse. The authors introduce an artificial neural network approach to voltage security monitoring and control. The neural network uses its association mechanism to approximate the complicated mathematical formulation of the voltage collapse phenomenon. The inherent parallel information processing nature of the neural network, which provides the capability of fast computation, enables the neural network approach to meet the rigorous demands of real-time monitoring and control. The IEEE 57 busbar system is used to demonstrate the applicability of the artificial neural network approach to the problem of voltage security monitoring and control in power systems View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural network based preventive control support system for power system stability enhancement

    Publication Year: 1991 , Page(s): 149 - 153
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (368 KB)  

    The authors propose an application of a newly developed neural network to the preventive control of a power system. The purpose of the proposed control is to improve the damping effect of the system on electromechanical modes by reallocating load to generators. Since the neural network has flexible learning capability the authors apply it to identify the complex and nonlinear relation between the damping effect and the distribution of generating power. The trained neural network acts as the support system which aids an operator in performing the generating reallocation for enhancing the system stability. Furthermore, the authors develop a new type of neural network which can deal with the equal constraints about the output layer in the error-back-propagation type of neural network because it is important for the generating reallocation to satisfy the equal constraint about the energy balance between generation and load View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Artificial neural network & pattern recognition approach for narrowband signal extraction

    Publication Year: 1991 , Page(s): 288 - 292
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (360 KB)  

    Estimation of unknown frequency, extraction of narrowband signals buried under noise and periodic interference are accomplished by employing existing techniques. However, the authors propose an artificial neural net based scheme together with pattern classification algorithm for narrowband signal extraction. A three layer feedforward net is trained with three different algorithms namely backpropagation, Cauchy's algorithm with Boltzmann's probability distribution feature and the combined backpropagation-Cauchy's algorithm. A constrained tangent hyperbolic function is used to activate individual neurons. Computer simulation is carried out with inadequate data to reinforce the idea of the net's generalization capability. The robustness of the proposed scheme is claimed with the results obtained by making 25% links faulty between the layers. Performance comparison of the three algorithms is made and the superiority of the combined backpropagation-Cauchy's algorithm is established over the other two algorithms. Simulation results for a wide variety of cases are presented for better appraisal View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dynamical implications of using neural networks as controllers

    Publication Year: 1991 , Page(s): 134 - 138
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (296 KB)  

    The authors examine the usefulness of the feedforward neural network as a controller. For illustrative purposes, the authors consider the case of controlling two-dimensional linear systems. Observations are then made which generalize to higher dimensions and nonlinear systems. Examples are provided to verify the results. In particular, a classical power system stabilizer is examined to demonstrate the feasibility of using a neural controller View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fish identification from sonar echoes-preprocessing and parallel networks

    Publication Year: 1991 , Page(s): 183 - 187
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (236 KB)  

    Environmental regulations require Ontario Hydro to conduct a series of aquatic surveys to monitor fish population in the neighbourhoods of the generating stations. Studies are currently under way in an attempt to replace the current netting methods used for the survey with sonar based methods which will be nonconsumptive as well as less expensive. The authors look at the use of multi-layer perceptrons to identify the fish from their sonar echoes. The current phase of the work investigates the impact of preprocessing techniques and the use of networks in parallel on the generalization properties. It is found that significant improvements are possible using simple combinations of three-layer perceptrons which have been trained using outputs from different preprocessors. In the test case studied, over 93 percent of the targets were identified correctly by the network View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Unsupervised learning strategies for the detection and classification of transient phenomena on electric power distribution systems

    Publication Year: 1991 , Page(s): 107 - 111
    Cited by:  Papers (4)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (496 KB)  

    A number of utilities are currently installing high-speed data acquisition equipment in their distribution substations. This equipment will make it possible to record the transient waveforms due to events such as low and high-impedance faults, capacitor switching, and load switching. The authors describe the potential of applying unsupervised learning strategies to the classification of the various events observed by a substation recorder. Several strategies are tested using simulation studies and the effectiveness of unsupervised learning is compared to current classification strategies as well as supervised learning View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural net based correction of power system distortion caused by switching power supplies

    Publication Year: 1991 , Page(s): 198 - 202
    Cited by:  Papers (2)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (420 KB)  

    The correction of these power system distortions using neural networks is presented. A multi-layer neural network is trained (using error back propagation) to correct the distorted current waves. Based on the results obtained artificial neural network seems to offer a good solution for the important problem of correcting power system harmonic distortion View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An adaptively trainable neural network algorithm and its application to electric load forecasting

    Publication Year: 1991 , Page(s): 7 - 11
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (352 KB)  

    A training procedure that adapts the weights of a trained layered perceptron type artificial neural network to training data originating from a slowly varying nonstationary process is proposed. The resulting adaptively trained neural network (ATNN), based on nonlinear programming techniques, is shown to adapt to new training data that is in conflict with earlier training data with affecting the neural networks' response minimally to data elsewhere. The ATNN demonstrates improved accuracy over conventionally trained layered perceptron when applied to the problem of electric load forecasting View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural networks for topology determination of power systems

    Publication Year: 1991 , Page(s): 297 - 301
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (408 KB)  

    The authors describe a parallel distributed topology classifier. The idea is to determine the system configuration in a very fast way, even in the presence of incorrect or unavailable switch/breaker status and analog measurements. A new supervised learning algorithm suitable for very large training sets is introduced 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 a fuzzy engineering tool

    Publication Year: 1991 , Page(s): 36 - 40
    Cited by:  Papers (6)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (424 KB)  

    The authors describe a knowledge engineering tool for short-term load forecasting to be used as an aid in operation and planning of distribution systems. This engineering tool is composed by two parts. Firstly, an artificial neural network is trained to produce the first evaluation of forecasted load. Following, a fuzzy expert system manipulate actual and forecasted values of real power and weather conditions to find the final forecasted load. Illustrative examples are presented using Hydro-Quebec Power System data View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Application of neural networks in numerical busbar protection systems (NBPS)

    Publication Year: 1991 , Page(s): 117 - 121
    Cited by:  Papers (6)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (244 KB)  

    During the development of a (conventional) busbar protection algorithm which is able to cope with current signals distorted by current transducer saturation, the question came up, whether it would be possible to use a neural network for preprocessing the data and restoring the distorted signals. A training tool for neural networks and a set of typical distorted and undistorted current signals was selected for a verification of the idea. The test showed that the application of a neural network to this issue is possible in principal and that the signal quality is improved with respect to the needs of a busbar protection system, respectively. The ability of the neural networks to map an increasing number of input signals to reasonable output signals is investigated. Furthermore some studies were made for implementing the trained neural network in hardware View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Security assessment of a turbine generator using H control based on artificial neural networks and expert systems

    Publication Year: 1991 , Page(s): 49 - 53
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (424 KB)  

    The authors describe a preliminary framework for real time security assessment of turbine generators that integrates artificial neural networks (ANN) and knowledge-based expert systems (KBES). The authors also present the transient stability assessment of a turbine generator using a back propagation artificial neural network. Additional signals have been added to the AVR and governor loops of the turbine generator using H control. The ANN's ability to learn, interpolate and reproduce behaviour is presented, showing how the stability of a high order nonlinear system can be obtained without the prior solution of the state equations View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Back-propagation as the solution of differential-algebraic equations for artificial neural network training

    Publication Year: 1991 , Page(s): 242 - 244
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (232 KB)  

    The backpropagation algorithm for neural network training is formulated as the solution of a set of sparse differential algebraic equations (DAE). These equations are then solved as a function of time. The solution of the differential equations is performed using an implicit integrator with adjustable time step. The topology of the Jacobian matrix associated with the DAE's is illustrated. A training example is included View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Approximations of power system dynamic load characteristics by artificial neural networks

    Publication Year: 1991 , Page(s): 178 - 182
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (272 KB)  

    The static and dynamic characteristics of power system loads are critical to obtaining quality operating point predictions or stability calculations. The composite behavior of components at load buses are usually too complicated to be expressed in a simple form. Based on the approximation capability of artificial neural networks the authors explore the possibility of using neural networks to emulate load behaviours. The results verify the potential of load representation by neural networks View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Application of a revised Boltzmann machine to topological observability analysis

    Publication Year: 1991 , Page(s): 283 - 287
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (348 KB)  

    The author presents a method for determining power system topological observability with a stochastic neural network. The proposed method is based on the Boltzmann machine that can cope with stochastic behavior of neurons. The Boltzmann machine is useful for solving combinatorial problems since it can avoid local minima. In this paper, a revised Boltzmann machine is proposed to improve the convergence characteristics. A squashing function is utilized to decrease the number of neurons in handling the inequality constraints of the topological observability problem View full abstract»

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