Recognition of power quality events using DT-DWT based Complex Wavelet Transform

This paper presents the use of DT-DWT (Dual Tree-Discrete Wavelet Transform) based CWT (Complex Wavelet Transform) technique for detecting and localizing the power quality (PQ) events like sag, swell, interruption, harmonics, transients and flicker. CWT is the complex valued extension to the standard DWT (Discrete Wavelet Transform) which suffers from the limitations like shift sensitivity, poor directionality and the absence of the phase information. A data base of these events is generated in MATLAB from the numerical models of these events within the parameters as per IEEE-1159 standard. Various features like mean, standard deviation, skewness, kurtosis, energy, entropy etc. are extracted to detect PQ events. An ANN (Artificial Neural Network) technique is used as a classifier and the classification results are presented to demonstrate the efficacy of the DT-DWT based CWT.