A Fast Fault Detection and Identification Approach in Power Distribution Systems

In this paper, a Modified Multi-Class Support Vector Machines (MMC-SVM) technique is developed to simultaneously detect and classify different types of open-circuit faults in power distribution systems. This technique is capable of detecting and identifying open-circuit faults considering the impact of variations in the voltage of different nodes in power distribution systems. The RMS (Root Mean Square) voltage of the power grid is used as the input signal to diagnose the faults. Simulations are carried out on the IEEE 13-node test system considering temporary open-circuit faults in MATLAB software. The simulation results show the accuracy, effectiveness, and robustness of the proposed method.