신경회로망과 Classifier를 이용한 부분방전패턴의 인식

Recognition of Partial Discharge Patterns using Classifiers and the Neural Network

In th is work, two approaches were proposed for the recognition of partial discharge patterns . The first approach was neural net work with back propagation a lgorithm , and the second approach was an glecalculation between two operator vectors. P D sig n a ls were detected using three electrode systems ； IE C ( b ) , needle-planeana C IG R E method II electrode system . Both of neural network and an gle com parison method show ed good recognition perform ance for the patterns similarto the train e d patterns . And the number of operators to be used had a great influence on the recognition perform ance to the untrained patterns .