Impact of Mother Wavelet on the Performance of Wavelet-Neural Network (WNN) Based Transformer Protection

Abstract

In this article, impact of the mother wavelet on the main power transformer protection has been identified. Continuous wavelet transformation (CWT) is used as the protection on signal processing tool in this scheme. CWT has great potentials in analyzing system transients. CWT coefficients of fault current and inrush current's rms values are applied to the input of the neural. The logic signals are obtained which indicates if they are fault current or inrush current at the output of artificial neural network. The appropriate choice of mother wavelet while analysis of signal with wavelet transformation affects the protection of transformer significantly. Because ability to distinguish fault current and inrush current ranges from one mother wavelet to another mother wavelet. © 2016 The Chamber of Turkish Electrical Engineers.