Artificial Neural Networks (ANNs) and Artificial Intelligence (AI) methodologies are beginning to play a significant role in power systems research. A combination of ANN/AI methodologies can be forged into a formidable technique using the symbolic strengths of AI to aid the massively parallel and distributed processing models utilized by ANNs. The authors attempt to bring a philosophical perspective to this hybrid approach and examine it from different angles. Voltage stability enhancement is used as an example area and the ideas are being tested on it. The main objective is to promote discussion amongst the researchers and to investigate how this method can be used effectively.<>
{"title":"A hybrid artificial neural network/artificial intelligence approach for voltage stability enhancement","authors":"S. Vadari, S. Venkata","doi":"10.1109/ANN.1991.213514","DOIUrl":"https://doi.org/10.1109/ANN.1991.213514","url":null,"abstract":"Artificial Neural Networks (ANNs) and Artificial Intelligence (AI) methodologies are beginning to play a significant role in power systems research. A combination of ANN/AI methodologies can be forged into a formidable technique using the symbolic strengths of AI to aid the massively parallel and distributed processing models utilized by ANNs. The authors attempt to bring a philosophical perspective to this hybrid approach and examine it from different angles. Voltage stability enhancement is used as an example area and the ideas are being tested on it. The main objective is to promote discussion amongst the researchers and to investigate how this method can be used effectively.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"115 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124813826","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
H. Sasaki, J. Kubokawa, M. Watanabe, R. Yokoyama, R. Tanabe
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.<>
{"title":"A solution of generation expansion problem by means of neutral network","authors":"H. Sasaki, J. Kubokawa, M. Watanabe, R. Yokoyama, R. Tanabe","doi":"10.1109/ANN.1991.213474","DOIUrl":"https://doi.org/10.1109/ANN.1991.213474","url":null,"abstract":"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.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"183 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123448154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The authors propose a new method for artificial neural networks (ANNs) based steady state equivalents of power systems. Because the multilayer Perceptron network (MPN) is a typical ANN and its training algorithm is quite effective, the authors use this network. When the studied power system is divided into three parts, which are internal system (IS), external system (ES) and boundary system (BS). Some tests show that the method has advantages of high accuracy, powerful suitability and high recognition speed.<>
{"title":"Artificial neural networks based steady state equivalents of power systems","authors":"Y. Jilai, L. Zhuo","doi":"10.1109/ANN.1991.213483","DOIUrl":"https://doi.org/10.1109/ANN.1991.213483","url":null,"abstract":"The authors propose a new method for artificial neural networks (ANNs) based steady state equivalents of power systems. Because the multilayer Perceptron network (MPN) is a typical ANN and its training algorithm is quite effective, the authors use this network. When the studied power system is divided into three parts, which are internal system (IS), external system (ES) and boundary system (BS). Some tests show that the method has advantages of high accuracy, powerful suitability and high recognition speed.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122383502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The high speed desired in the implementation of many neural network algorithms, such as backpropagation learning in a multilayer perceptron (MLP), may be attained through the use of finite precision hardware. This finite precision hardware, however, is prone to errors. A method of theoretically deriving and statistically evaluating this error is presented and could be used as a guide to the details of hardware design and algorithm implementation. The paper is devoted to the derivation of the techniques involved as well as the details of the backpropagation example. The intent is to provide a general framework by which most neural network algorithms under any set of hardware constraints may be evaluated.<>
{"title":"Finite precision error analysis for neural network learning","authors":"J. L. Holt, Jenq-Neng Hwang","doi":"10.1109/ANN.1991.213471","DOIUrl":"https://doi.org/10.1109/ANN.1991.213471","url":null,"abstract":"The high speed desired in the implementation of many neural network algorithms, such as backpropagation learning in a multilayer perceptron (MLP), may be attained through the use of finite precision hardware. This finite precision hardware, however, is prone to errors. A method of theoretically deriving and statistically evaluating this error is presented and could be used as a guide to the details of hardware design and algorithm implementation. The paper is devoted to the derivation of the techniques involved as well as the details of the backpropagation example. The intent is to provide a general framework by which most neural network algorithms under any set of hardware constraints may be evaluated.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114484513","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the proliferation of sensitive control systems and personal computers in the commercial and industrial sector, comes a need for electrical utilities to deliver 'clean' power. Voltage variations in the form of sags, surges and impulses, i.e., disturbances, can chronically plague and permanently damage electrical equipment. Southern California Edison (SCE) in joint effort with Basic Measuring Instruments (BMI) were teamed up to automate the process of collecting disturbance data, viewing their contents and applying artificial intelligence paradigms (neural networks) to help identify their causes and present possible solutions.<>
{"title":"Power quality monitoring using neural networks","authors":"R.F. Daniels","doi":"10.1109/ANN.1991.213479","DOIUrl":"https://doi.org/10.1109/ANN.1991.213479","url":null,"abstract":"With the proliferation of sensitive control systems and personal computers in the commercial and industrial sector, comes a need for electrical utilities to deliver 'clean' power. Voltage variations in the form of sags, surges and impulses, i.e., disturbances, can chronically plague and permanently damage electrical equipment. Southern California Edison (SCE) in joint effort with Basic Measuring Instruments (BMI) were teamed up to automate the process of collecting disturbance data, viewing their contents and applying artificial intelligence paradigms (neural networks) to help identify their causes and present possible solutions.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128253977","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The authors address the problem of training neural networks to act as approximations of continuous mappings. In the case where the only representation of the mapping within the training process is through a finite set of training points, they show that in order for this set of points to provide an adequate representation of the mapping, it must contain a number of points which rises at least exponentially quickly with the dimension of the input space. Thus they also show that the time taken to train the networks will rise at least exponentially quickly with the dimension of the input. They conclude that if the only training algorithms available rely upon a finite training set, then the application of neural networks to the approximation problem is impractical whenever the dimension of the input is large. By extrapolating their experimental results, they estimate that 'large' in this respect means 'greater than ten'.<>
{"title":"On the number of training points needed for adequate training of feedforward neural networks","authors":"K. Hashemi, R. J. Thomas","doi":"10.1109/ANN.1991.213472","DOIUrl":"https://doi.org/10.1109/ANN.1991.213472","url":null,"abstract":"The authors address the problem of training neural networks to act as approximations of continuous mappings. In the case where the only representation of the mapping within the training process is through a finite set of training points, they show that in order for this set of points to provide an adequate representation of the mapping, it must contain a number of points which rises at least exponentially quickly with the dimension of the input space. Thus they also show that the time taken to train the networks will rise at least exponentially quickly with the dimension of the input. They conclude that if the only training algorithms available rely upon a finite training set, then the application of neural networks to the approximation problem is impractical whenever the dimension of the input is large. By extrapolating their experimental results, they estimate that 'large' in this respect means 'greater than ten'.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123207553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Sendaula, S. Biswas, A. Eltom, C. Parten, W. E. Kazibwe
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.<>
{"title":"Application of artificial neural networks to unit commitment","authors":"M. Sendaula, S. Biswas, A. Eltom, C. Parten, W. E. Kazibwe","doi":"10.1109/ANN.1991.213467","DOIUrl":"https://doi.org/10.1109/ANN.1991.213467","url":null,"abstract":"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.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123725173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The ability of a recurrent network to model load forecasting is investigated. Its performance in a competition is then contrasted with that of feedforward networks and linear models. Its weaknesses and strengths are then analyzed to give guidelines to the design of neural net predictors with the hope of designing better predictors in the future.<>
{"title":"Recurrent neural networks and load forecasting","authors":"J. Connor, L. Atlas, D. Martin","doi":"10.1109/ANN.1991.213491","DOIUrl":"https://doi.org/10.1109/ANN.1991.213491","url":null,"abstract":"The ability of a recurrent network to model load forecasting is investigated. Its performance in a competition is then contrasted with that of feedforward networks and linear models. Its weaknesses and strengths are then analyzed to give guidelines to the design of neural net predictors with the hope of designing better predictors in the future.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130903452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The authors propose a method by which a set of artificial neural networks (ANNs) can be used as a dispatchers' aid in alarm processing. The proposed model consists of two parts-a rule based system and a set of neural networks. Depending on the input that is fed to the system, a set of rules that make up the rule based system, are used to activate one or more ANNs. Each ANN in the system is used to identify the faults in a single sub-station or a particular zone. The rule based system is also used to aid the ANNs in identifying multiple faults, by activating them in the required order and providing them with the necessary alarms as inputs.<>
{"title":"Artificial neural networks as a dispatcher's aid in alarm processing","authors":"R. Karunakaran, G. Karady","doi":"10.1109/ANN.1991.213484","DOIUrl":"https://doi.org/10.1109/ANN.1991.213484","url":null,"abstract":"The authors propose a method by which a set of artificial neural networks (ANNs) can be used as a dispatchers' aid in alarm processing. The proposed model consists of two parts-a rule based system and a set of neural networks. Depending on the input that is fed to the system, a set of rules that make up the rule based system, are used to activate one or more ANNs. Each ANN in the system is used to identify the faults in a single sub-station or a particular zone. The rule based system is also used to aid the ANNs in identifying multiple faults, by activating them in the required order and providing them with the necessary alarms as inputs.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133796895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The authors discuss the use of supervised learning and associative memories in an application for protecting the power system during an emergency situation. Automatic devices based on artificial neural networks are proposed as an intelligent and fast tool to mitigate the consequences of the major disturbance in the power system, area that involves a lot of unsolved problems. To prove the concept, the artificial neural network was trained to perform generation rescheduling as a way to alleviate the line overloads. The IEEE-30 bus test system was used to demonstrate that a feedforward neural network with back propagation can detect the state of the power system by monitoring line flows from SCADA data and then, make recommended corrective actions.<>
{"title":"Identification of power system emergency actions using neural networks","authors":"D. Novosel, R. King","doi":"10.1109/ANN.1991.213477","DOIUrl":"https://doi.org/10.1109/ANN.1991.213477","url":null,"abstract":"The authors discuss the use of supervised learning and associative memories in an application for protecting the power system during an emergency situation. Automatic devices based on artificial neural networks are proposed as an intelligent and fast tool to mitigate the consequences of the major disturbance in the power system, area that involves a lot of unsolved problems. To prove the concept, the artificial neural network was trained to perform generation rescheduling as a way to alleviate the line overloads. The IEEE-30 bus test system was used to demonstrate that a feedforward neural network with back propagation can detect the state of the power system by monitoring line flows from SCADA data and then, make recommended corrective actions.<<ETX>>","PeriodicalId":119713,"journal":{"name":"Proceedings of the First International Forum on Applications of Neural Networks to Power Systems","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116442919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：