An approach to classify transient disturbances with incomplete S-transform and Morlet wavelet spectral kurtosis

To classify and recognize the transient power quality disturbances (PQDs), an algorithm based on incomplete S-transform (IST) and Morlet wavelet spectral kurtosis (SK) is proposed. Firstly, Morlet wavelet SK and IST are used to obtain characteristics of different transient PQDs. Characteristics extracted from the IST and the Morlet wavelet SK are respectively used to classify amplitude disturbances and disturbances of oscillation and impulsive transient. Then the K-means algorithm is used to give a clustering center of the classification, and different disturbances are distinguished from each other by different clustering center. Simulation results show that the transient disturbance characteristics can be effectively extracted with the presented algorithm.