On the choice of wavelet based features in power quality disturbances classification

In this study we investigate the effectiveness of some wavelet based features used for classification of power quality (PQ) disturbances, and check the difference in their efficiency when they are used in combinations, in order to perform optimal wavelet based feature extraction method. The investigation was made using three well known classification techniques, which are support vector machine (SVM), decision tree (DT) and random forest (RF), for classification in case of 7 and 11 types of PQ disturbances. In both cases it is shown that the effectiveness of a given feature is not general, but it depends on the type of other features it is used with and the kind of the applied classification method.