On the use of wavelets for the detection and analysis of power system transients

This paper discusses how the powerful wavelet analysis technique is applied to the detection and analysis of power system transients. Although traditional Fourier analysis techniques are suitable for periodic signals, the analysis of aperiodic signals, caused by a transient event, benefit from the robust wavelet analysis technique. The basic idea is to decompose the signal into a number of detailed and smoothed versions, which contain the high-and low-frequency components of the signal, respectively. By analyzing the detailed version of the signal at different levels of resolution, both time and frequency localization of the event can be achieved. Consequently, the wavelet analysis technique can focus on both the very short time interval and the high frequency components of the power system transient so as to enhance its detection and characterization.