Power quality analysis through spike detection using smoothed pseudo Wigner-Ville distribution

The extensive usage of nonlinear electrical loads resulted in increased susceptibility to disturbances related to the quality of power(voltage sag, voltage swell, spike, oscillatory-transients etc) in the electric-power system. Therefore, the power quality disturbances are to be analyzed to ensure reliable power distribution through decentralized generation system. This paper deals with the detection of spike using joint time-frequency analysis method based on smoothed pseudo Wigner-Ville distribution as it has highest concentration of signal energy in time-frequency plane with reduction in cross-terms. The smoothed pseudo Wigner-Ville distribution (SPWVD) is a quadratic time-frequency distribution, which has been recently proposed in the literature, due to its inherent characteristics for analyzing non-stationary signals. To the best of authors' knowledge smoothed pseudo Wigner distribution has not used for electrical spike detection in power systems. For the of experimentation, the test signals have been generated using MATLAB software and tested in both clean and noisy environments. The performance of the proposed proposed is found to be comparable with short-time Fourier transform, and S-transform based techniques.