A Data-Driven-Based Wide-Area Protection Scheme for Fault Detection Using the Limited Measurements

This paper presents a novel and efficient approach for wide-area fault detection in microgrids, utilizing data-driven techniques based on voltage and current measurements. The proposed method offers both high speed and accuracy in detecting faults. The methodology consists of three key steps that collectively form a comprehensive protection scheme. Initially, the current trajectories obtained from the measurements are analyzed to determine the fault condition. This initial indicator serves as a valuable starting point for fault detection. In the second step, the impedance of the lines, including the considered area, is calculated for the fault detection. The change of the calculated impedances implies for the fault occurrence activating the third step. In the final step, an iterative process is followed to identify the faulted line. The proposed method provides a faster and more reliable fault detection mechanism, allowing for rapid response and mitigation of potential disruptions. The efficacy of the proposed method is validated on an 11-bus microgrid. The simulation investigations are conducted in MATLAB\SIMULINK environment.