A feed-forward artificial neural network with enhanced feature selection for power system transient stability assessment

Sawhney, Harinder; Jeyasurya, B.

doi:10.1016/j.epsr.2005.12.026

Cited by 61 publications

(34 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Trata-se de um procedimento bastante útil para a resolução de problemas abordados na operação (Chicco et al, 2005;Levi et al 2005), bem como no planejamento. Ou seja, visa reduzir o volume de processamento e a dimensão do conjunto de dados para o treinamento e análises, quando se emprega, por exemplo, outras metodologias neurais para a resolução de problemas como: análise de estabilidade transitória (Ferreira et al, 2006;Sawhney et al 2006), análise de estabilidade de tensão, previsão de cargas elétricas, etc. Aplicamse, ainda, quando há necessidade de conhecer características de similaridade, dentro de um conjunto de dados ou padrões.…”

Section: Introductionunclassified

Formação De Classes De Pefis De Geração E De Carga Em Sistemas Elétricos Usando Uma Rede Neural Nebulosa

Trindade¹,

Minussi²

2016

Anais Do 8. Congresso Brasileiro De Redes Neurais

View full text Add to dashboard Cite

Abstract This work presents a methodology for generation and load profiles clustering formation based on fuzzy-ART (Adaptive Resonance Theory) neural network. It is a method that seeks to form the similar groups of nodal electric power, with the purpose of reducing the computacional effort used in the resolution of a series of studies accomplished in electric power systems, e.g., transient stability analysis, voltage stabiliy analysis, distribution system planning, etc. The ART architectures present plasticity and stability characteristics, which are very important for the training and to execute the analysis in a fast way, allowing the continuous training. The continuous training consists in the inclusion the new patterns, if availables, in an appropriate way, on the weight matrix of the ART neural network. This was the main reason of the use this neural network ART-family. To illustrate the proposed neural network an application is presented for a multi-machine electric power systems composed of 10 synchronous machines, 45 buses and 73 transmission lines.Keywords Electric Power Systems, Clustering Technique, Nodal Electric Power Injection, Neural Networks, Adaptive Resonance Theory.Resumo Este artigo apresenta uma metodologia para a formação de classes de perfis de geração e de carga em sistemas de ene rgia elétrica baseada numa rede neural ART (Adaptive Resonance Theory) nebulosa. Trata-se de um procedimento que visa agrupar os vetores referentes às injeções de potência nodais por conjuntos similares, com o propósito de reduzir o esforço computacional empregado na resolução de uma série de estudos realizados em sistemas elétricos de potência, e.g., análise de estabilidade transitória, análise de estabilidade de tensão, planejamento de sistemas de distribuição, etc. As arquiteturas ART apresentam as características de estabilidade e de plasticidade, as quais são imprescindíveis para a realização do treinamento e execução da análise de forma rápida e continuada, ou seja, toda vez que houver disponibilidade de novos padrões, não haverá necessidade de reiniciar o treinamento, bastando apenas armazenar tais padrões, de forma adequada, na memória da rede neural (conjunto de pesos). Esta foi a principal razão do emprego desta rede neural da família ART-descendente. Como forma de ilustrar a estrutura neural proposta, apresenta-se uma aplicação considerando-se um sistema elétrico composto por 10 máquinas síncronas, 45 barras e 73 linhas de transmissão. Palavras-chave

show abstract

Section: Introductionunclassified

Formação De Classes De Pefis De Geração E De Carga Em Sistemas Elétricos Usando Uma Rede Neural Nebulosa

Trindade¹,

Minussi²

2016

Anais Do 8. Congresso Brasileiro De Redes Neurais

View full text Add to dashboard Cite

show abstract

“…Power systems' operators are required to continuously monitor the security of power systems for a probable set of contingency and be prepared to take appropriate preventive and emergency control measures if need arises. There are several established methods and software tools to evaluate transient security of power system for a probable set of contingency [1][2][3][4][5]. During normal operation if an operating state is found to be insecure, preventive control is initiated to bring the system back to normal operating state.…”

Section: Introductionmentioning

confidence: 99%

Preventive and Emergency Control of Power System for Transient Stability Enhancement

Siddiqui¹,

Verma²,

Niazi³

et al. 2015

Journal of Electrical Engineering and Technology

View full text Add to dashboard Cite

-This paper presents preventive and emergency control measures for on line transient stability (security) enhancement. For insecure operating state, generation rescheduling based on a real power generation shift factor (RPGSF) is proposed as a preventive control measure to bring the system back to secure operating state. For emergency operating state, two emergency control strategies namely generator shedding and load shedding have been developed. The proposed emergency control strategies are based on voltage magnitudes and rotor trajectories data available through Phasor Measurement Units (PMUs) installed in the systems. The effectiveness of the proposed approach has been investigated on IEEE-39 bus test system under different contingency and fault conditions and application results are presented.

show abstract

“…Using a large number of features has two potential adverse effects. The time complexity of training a neural network increases dramatically with the increasing number of input features [15]. In addition, the existence of irrelevant features increases the number of inputs without providing new information.…”

Section: Introductionmentioning

confidence: 99%

“…To mitigate these adverse effects on the training of ANN and to increase its prediction performance, which directly affects the success in DSA, an effective feature selection must be applied. Various feature selection methods adopting the measures of feature quality such as sensitivity index [15], divergence [16] and Fisher discrimination [17] have been used for the DSA of power systems based on MLPs. In this study, the effect of using filter type feature selection methods, such as Minimum Redundancy Maximum Relevance (mRMR) [18] and Regressional ReliefF (RReliefF) [19], on the final performance of MLP for predicting the power system security indices is investigated.…”

Section: Introductionmentioning

confidence: 99%

A Novel Feature Selection Method for the Dynamic Security Assessment of Power Systems Based on Multi-Layer Perceptrons

Beyranvand

Kucuktezcan

Çataltepe

et al. 2018

ijisae

View full text Add to dashboard Cite

In this study, the effect of feature selection methods on the performance of multi-layer perceptrons used for the dynamic security assessment of electric power systems is investigated. The existence of many measurable parameters (features) characterizing the power system security status complicates the use of multi-layer perceptron both in terms of prediction accuracy and training time. In this paper, the dynamic security of a power system subject to a number of critical contingencies is assessed as the critical clearing time of any credible fault is predicted by a multi-layer perceptron. In addition to the study of two different feature selection methods, which are Minimum Redundancy Maximum Relevance (mRMR), and Regressional ReliefF (RReliefF), a novel multi-layer perceptron based feature selection method is proposed to be applied in the prediction of security indices. The performance of the feature selection methods on the dynamic security assessment is investigated on a 16-generator, 68-bus test system.

show abstract

A feed-forward artificial neural network with enhanced feature selection for power system transient stability assessment

Cited by 61 publications

References 7 publications

Formação De Classes De Pefis De Geração E De Carga Em Sistemas Elétricos Usando Uma Rede Neural Nebulosa

Formação De Classes De Pefis De Geração E De Carga Em Sistemas Elétricos Usando Uma Rede Neural Nebulosa

Preventive and Emergency Control of Power System for Transient Stability Enhancement

A Novel Feature Selection Method for the Dynamic Security Assessment of Power Systems Based on Multi-Layer Perceptrons

Contact Info

Product

Resources

About