Fault Detection in DC Microgrids Using Short-Time Fourier Transform

Fault detection in microgrids presents a strong technical challenge due to the dynamic operating conditions. Changing power generation and load impacts current magnitude direction, which has an adverse effect on microgrid protection scheme. To address this problem, paper addresses field-transform-based fault method immune The faults are simulated via Matlab/Simulink model of grid-connected photovoltaics-based DC with battery energy storage. Short-time Fourier transform is applied time signal obtain frequency spectrum. Selected spectrum features then provided number intelligent classifiers. classifiers’ scores were evaluated using F1-score metric. Most classifiers proved be reliable as their performance score was above 90%.