Method for classification of PQ events based on discrete Gabor transform with FIR window and T2FK-based SVM and its experimental verification

This paper aims to develop a new method for identification of Power Quality (PQ) events based on Discrete Gabor Transform (DGT) with a Finite Impulse Response Window (FIR-DGT) and Type-2 Fuzzy Kernel-based Support Vector Machine (T2FK-SVM). The FIR-DGT extracted features from the input signals and the T2FK-SVM classified them. Using proper window function for DGT is essential. Iterated sine window was used as window function to extract the events' features. The iterated sine window's function is four times faster than the default window function of DGT. Kernel design is a main part of many Kernel-based methods such as SVM, so by using T2FK, the total accuracy of classification is enhanced. In the present work, use of this hybrid approach decreased the extracted features size, so the execution time and total required memory were optimized for the classification section. The simulation results revealed accurate classification and execution in the detection and classification of nine types of PQ events. The overall accuracy of the proposed method was comparable to other methods and the accuracy evaluated under noisy conditions. Hardware platform was developed for evaluating the proposed method based on ARM LPC 1768 microcontroller to assess accuracy of the method in real conditions.