Cross Hilbert-Huang transform based feature extraction method for multiple PQ disturbance classification

This paper presents a new methodology of Cross-Hilbert Huang transform based feature selection for sensing simultaneous occurrence of multiple power quality disturbances. Kernel PCA is used for feature selection because this method is well suited for non-linear and non-stationary multiple power quality disturbances. A linear support vector machine is used for classification of the extracted features. Results show that the performance is comparable with the results reported in the literatures. The present method is generic in nature and can be applicable for topologically similar problems.