Differential Evolution-Based Matched Wavelet for Differential Protection of Transformer

Abstract

The work proposes a matched wavelet method for detection and discrimination of inrush and fault waveform in power transformers. The method rests on the concept that if analyzing wavelet is specialized to have shape matching with the waveform being analyzed (fault or inrush) its discrimination ability can improve drastically. The matched wavelet for inrush and fault waveforms is developed using Differential Evolution (DE) algorithm. It is established that the proposed method has sub-cycle (nearly half-cycle) discrimination ability. The method was tested by generating waveforms for all the angles to test its discrimination ability. © 2019, Springer Nature Singapore Pte Ltd.