A classification method of power quality disturbance based on wavelet packet decomposition

The power quality problems has been given more attention due to the proliferation of sensitive equipments. In order to mitigate the influence, various power quality disturbances must be classified before an appropriate action can be taken. Wavelet packet is developed on wavelet transform, which can provide more time-frequency information. This paper selects the energy and entropy of terminal nodes through wavelet packet decomposition as a feature vector respectively, using the Fisher linear classifier to design piecewise linear classifier in order to classify the disturbances, which are simulated and analyzed. The simulation results indicate that the entropy feature vector has a higher recognition accurate ratio.