Support Vector Machine for power quality disturbances classification using higher-order statistical features

Support Vector Machine (SVM), which is based on Statistical Learning theory, is a universal machine learning method. This paper proposes the application of SVM in classifying to several power quality disturbances. For this purpose, a process based in HOS has been realized to extract features that help in classification. In this stage the geometrical pattern established via higher-order statistical measurements is obtained, and this pattern is function of the amplitudes and frequencies of the power quality disturbances associated to the 50-Hz power-line. Once the features are managed will be segmented to form training and test sets and them will be applied in the statistical method used to perform automatic classification of PQ disturbances. The result is shown according to correlation and mistake rates.