DefinePK

Abstract

This paper proposes a novel approach for fault discovery of three-phase transmission line, which is primarily based upon the Discrete Wavelet Transformation (DFT) and Back propagation neural network algorithm. Three phase currents of most effective one end are measured, and discrete features are extracted using Discrete Wavelet Transform. Those features are then used as inputs to the Back propagation neural network algorithm. The Training data set for Back propagation neural network Algorithm is obtained by way of simulating the ten extraordinary kinds of faults the use of diverse values of fault inception angles and fault resistances, so that the actual consequences may be received. The proposed Back propagation neural network algorithm employs twenty inputs and best one output for classifying the faults. The distinctiveness of the proposed approach is that all the features, data used in developing the Algorithm are normalized, so that the method may be used for any system with none significant changes. The simulation of the Three-phase transmission line network and discrete Wavelet Transformation analysis are achieved inside the toolboxes of MATLAB®, and Back propagation neural network Algorithm codes also are written in MATLAB®.