A Fuzzy-Neural Technique for Flashover Diagnosis of Winding Insulation in Transformers

Abstract

A Fuzzy-Neural pattern recognition technique for insulation flashover diagnosis in oil filled power transformers has been described in the paper. Determination of exact nature and location of internal insulation flashover during impulse testing of power transformers is of practical importance to the manufacturer as well as designers. The presently available diagnostic techniques more or less depend on expertise of the test personnel and in many cases not beyond ambiguity and controversy. The new technique described in the paper relies on high discrimination power and excellent generalization ability of Fuzzy-Neural systems in complex pattern classification problem and overcomes the limitations of conventional diagnostic methods. The technique applied to winding model of typical high voltage transformers exhibited high diagnostic accuracy by successful detection and discrimination of flashovers of different nature and site of occurrence in the winding.KeywordsPower TransformerSpectrum PatternFuzzy ARTMAPTraining ExemplarVigilance ParameterThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.