Detection and Localization of Series Arc Faults in DC Microgrids Using Kalman Filter

DC networks are becoming more popular in a wide range of applications. However, the difficulty in detecting and localizing a high-impedance series arc fault presents, a major challenge slowing the wider deployment of dc networks/microgrids. In this article, a Kalman filter (KF)-based algorithm to monitor the operation of a dc microgrid by estimating the line admittances and consequently detecting/localizing series arc faults is introduced. The proposed algorithm uses the voltage and current samples from the nodes in the distribution network to estimate the line admittances. By determining these values, it is possible to quickly isolate the faulted section and reconfigure the network after a fault occurs. Since the disturbance caused by a high-impedance series arc fault spreads across almost the entire microgrid, the KF algorithm is structured to detect the faulted line in the grid with precision. Simulation and control hardware-in-the-loop (CHIL) results are presented, demonstrating the feasibility of implementation.

Automated summary

DC networks are increasingly popular, but detecting and locating high-impedance series arc faults remains a major challenge. A Kalman filter-based algorithm is introduced in this article to monitor DC microgrid operation by estimating line admittances and detecting/localizing series arc faults. The algorithm uses voltage and current samples to estimate line admittances, isolate faulted sections, and reconfigure the network quickly after a fault. Simulation and control hardware-in-the-loop results demonstrate the feasibility of implementation.