Fault Detection and Classification in Transmission Lines Using Wavelet Transform and Support Vector Machines

Abstract

This paper proposes a method using Discrete Wavelet Transform and Support Vector Machines (SVM) to detect and classify faults along a transmission line. Discrete Wavelet Transform is used to detect the high frequency components contained in a fault signal spectrum. Different faults (short circuits) on different positions of a transmission line are simulated using a link between MATLAB and ATP/EMTP programs. The algorithm implemented in MATLAB is tested in various case studies based on the largest transmission line of the IEEE 9 bus system.