Monitoring classifier for power quality discrimination using wavelet-grey relational analysis

This paper proposes a method of power quality (PQ) discrimination for power system using wavelet-grey relational analysis (WGRA). The monitoring classifier based on WGRA can be divided into two stages, Gaussian wavelets are used to extract the features from distorted waves and reconstruct various patterns, and grey relational analysis (GRA) discriminates the disturbance events. The proposed monitoring classifier was used to test for the power quality disturbances, including those caused by harmonics, voltage sag, voltage swell, and voltage interruption. Compared with the wavelet networks, the test results will show accurate discrimination, good robustness, and faster processing time for detecting disturbing events.