Transmission Lines Fault Detection, Classification and Location Considering Wavelet Support Vector Machine with Harris Hawks Optimization Algorithm to Improve the SVR Training

Abstract

This research presents a novel synthetic framework which can efficiently detect the short circuit faults, classify them, and find their location in transmission lines. The suggested approach relies on the measured voltage and current waveforms when faults occur. To detect the faults, discrete wavelet transform (DWT) needs to be implemented to the measured currents. When a fault happens, the classification module is activated by employing the support vector machine (SVM) technique and DWT. To determine the fault location accurately, Harris Hawks optimization (HHO) algorithm is utilized for improving the SVR training process. In this study, fault data samples are created based on the fault type, location changes, and ground resistance. The case study network was modelled in PSCAD/EMTDC and MATLAB. The outcomes illustrate the precision and efficacy of the suggested framework.