Spectral energy function for fault detection during power swing

Abstract

This paper presents a spectral energy function for fault detection during power swing using a novel time frequency transform known as S-transform. S-transform is a variable window short time Fourier Transform (STFT) which combines the elements of STFT and Wavelet transform and, thus known as phase corrected Wavelet transform. The process starts at processing the current signal through S-transform to generate the S-matrix. From the S-matrix, corresponding S-contours (time-frequency contours) are generated. The spectral energy content of the S-counters is used to register the symmetrical and un-symmetrical faults during power swing. The proposed technique is tested for different fault conditions during power swing with possible variations in operating parameters and is able to identify the faults with a response time of 1.25 cycles from the fault inception during power swing.