Feature vector extraction for the automatic classification of power quality disturbances

The objective of this paper is to present a systematic approach to feature vector extraction for the automatic classification of power quality (PQ) disturbances, where discrete wavelet transform (DWT), signal power estimation and data compression methods are utilized to improve the classification performance and reduce computational complexity. To demonstrate the performance and applicability of the proposed method, some test results obtained by analyzing 7-class power quality disturbances, generated by the EMTP, with white Gaussian noise are also provided.