Método de Classificação de Faltas em Linhas de Transmissão baseado em Estatísticas de Ordem Superior

Abstract

Resumo-It is known that the study of power transmission lines has been the subject of several researches aiming to provide relevant information to users of electrical systems. This study provides an approach for fault classification using higher order statistics and a neural network based classifier. A detector based on euclidean distance was implemented to reduce the complexity of the classifier. This implementation enables real-time execution because only 1 ⁄32 cycles of post-fault data are used in feature extraction. It was able to classify 10 classes of faults with global efficiency upper to 97%.