Fault Classification and Ground detection using Support Vector Machine

Abstract

A new approach for fault detection in power system network using time-frequency analysis is presented in this paper. The S-transform with complex window is used for generating frequency contours(S-contours), which distinguishes the faulted condition from no-fault. Here the fault current data for one cycle back and one cycle from the fault inception is processed through S-transform to generate time-frequency patterns with varying window. The generated time-frequency patterns clearly distinguishes the faulted condition from un-faulted.