Analysis of composite power quality events using s-transform

Abstract

This paper demonstrates the effectiveness of S-transform in detecting and classifying multiple/composite power quality (PQ) events compared to the discrete wavelet transform for the accuracy of tracking the amplitude and frequency changes. Five different types of multiple/composite disturbance signals are considered, in this paper, including harmonics with voltage sag, harmonics with voltage swell, impulsive transient with harmonics, harmonics with notches, harmonics with voltage sag corrupted with noise. Analysis of different multiple PQ events to extract the quantitative features like time, frequency and amplitude of each disturbance clearly shows that the S-transform method is insensitive to noise and proves to be more accurate and effective as compared to wavelet transform.