Multi-class power quality disturbances classification by using ensemble empirical mode decomposition based SVM

This paper presents performance comparisons of Support Vector Machine (SVM) and different classification method for power quality disturbance classification. The first goal of this study is to investigate EEMD (ensemble empirical mode decomposition) performance and to compare it with classical EMD for feature vector extraction and selection of power quality disturbances. Features are extracted from the power electrical signals by using Hilbert Huang Transform (HHT). This technique is a combination of ensemble empirical mode decomposition (EEMD) and Hilbert transform (HT). The outputs of HT are instantaneous frequency (IF) and instantaneous amplitude (IA). Characteristic features are obtained from first IMFs´, IF and IA. The ten features, i.e. mean, standard deviation, singular values, maxima and minima of IF and IA, are then calculated. These features are normalized and the inputs of SVM and other classifiers.