Dual-tree complex wavelet packet transform and regularized extreme learning machine-based feature extraction and classification of power quality disturbances

There is an urgency to monitor, detect and classify the power quality disturbances (PQDs) due to global concern of the exponential increase of power quality (PQ) and the penetration of renewable energy sources (RESs) through power electronics-based interfacing devices to the power system. In this study, an intelligent method based on dual-tree complex wavelet packet transform (DTCWPT) is used to extract the frequency-domain features from the frequency-band signals to reveal PQ characteristics. To classify the PQDs, a regularized extreme learning machine (RELM) is used in which regularized regression method pruning the ELM by a computed suitable number of the hidden nodes in the network architecture. To test the efficiency of the proposed approach, the system is verified with eight PQDs with noise addition and multiple occurrence events cases and, at last, with practical signals generated from a PC interfaced hardware prototype. The proposed approach results in satisfactory performance for the PQ analysis and can be considered as the traditional PQ analyzer in real-time conditions.