Automatic classification of power quality disturbances: a review

Abstract

The development of intelligent power quality (PQ) disturbances classification and analysis tools exploited various digital signal-processing techniques to extract important features from the PQ signals. The purpose of this paper is to present a comprehensive review and discussion of the advanced tools for the automatic classification of PQ disturbances. The digital signal-processing tools applied for feature extraction include Fourier-transform, Wavelet-transform, Stockwell-transform etc. For the classification of PQ disturbances, the artificial intelligence techniques such as artificial neural networks, fuzzy logic and support vector machine are reviewed here. A large number of features used as inputs to the classifiers may affect the accuracy rate and requires a large memory space. The optimization techniques have been used in literature for optimal feature selection, which include genetic algorithm, simulated annealing, particle swarm optimization and ant colony optimization. An extensive review provides to the researchers a clear perspective on various techniques of PQ disturbances classification.