A new approach to time-frequency analysis and pattern recognition of non-stationary power signal disturbances

The paper presents a new approach to localize, detect and classify power signal disturbance problems using S-transforms (phase corrected wavelet transform) and a rule based expert system. The S-transform is an extension of the ideas of the continuous wavelet transform (CWT) and is based on a moving and scalable localizing gaussian window. This transform has some desirable characteristics that are absent in the continuous wavelet transform. The S-transform is unique in that it provides frequency-dependant resolution while maintaining a direct relationship with the Fourier spectrum. These advantages of the S-transform are due to the fact that the modulating sinusoids are fixed with respect to the time axis, whereas the localizing scalable gaussian window dilates and translates. Several power signal transient disturbances like voltage sag, voltage swell, flicker, harmonic distortions are taken for analysis using both S-transforms and wavelet transforms to prove the superiority of the former over the later. Automated classification software is developed using Artificial Intelligence technique like Expert Systems to provide very accurate identification of power quality events.