Power System Transient Analysis Using Wavelet Transform

This paper presents time-frequency multiresolution wavelet analysis on power system transients. Since it is difficult to analyze transients on-site due to a vast amount of data recorded from a digital transient recorder, it is imperative to develop an intelligent recorder which has the ability to detect and classify different types of power system transients. In order to develop a good approach for transient data capture and analysis, one of the important steps is to model transients, so that it is more efficient for computers to extract the features from power transients and recognize them automatically. Most of transient signals are aperiodic and short-term duration signals superimposed on the power frequency. The Fourier-based transforms have been proved valuable for periodic, time-invariant, or stationary signals, but they have problems with analyzing power transients due to the lack of multiple resolution in time-frequency domain. Wavelet analysis overcomes the limitations of the Fourier-based transforms by employing analyzing functions that are localized both in time and frequency domain [RCMG96]. Discrete Daubechies wavelet transforms with different vanishing moments [Daub92] have been used to investigate several simulated transients such as impulsive transients, oscillatory transients, voltage dip or swell, momentary interruption, and so on [HuHH99]. The transforms are accomplished via multiresolution signal decomposition (MSD) technique. The wavelet transform decomposes transients into a series of wavelet components, each of which corresponds to a timedomain signal that covers a specific octave frequency band containing more detailed information. Such wavelet components appear to be useful for detecting, localizing, and classifying the sources of transients. Moreover, the higher the vanishing moment, the more details of transients the wavelet transform can offer. In summary, the wavelet transform is feasible and practical for analyzing power system transients.