Discriminati on between external short circuit and internal winding fault in power transformer using discrete wavelet transform and back-propagation neural network

Abstract

This paper proposes an algorithm based on a combination of discrete wavelet transform (DWT) and back-propagation neural network (BPNN) for detecting and identif y ing internal winding fault of three-phase two-winding transformer. The maximum ratio obtained from division algorithm between coefficient from DWT of differential current and zero sequence for post-fault differential current waveforms is emplo y ed as an input for the training pattern in order to discriminate between internal fault and external short circuit. Variouscases studies based on Thailand electricit y transmission and distribution s y stems have been investigated so that the algorithm can be implemented. Results show that the proposed technique has good accurac y to detect fault and to identif y its position in the considered s y stem.