Arc Fault Detection in Photovoltaic Systems Based on Pseudo Wigner-Ville Distributed Algorithm

Abstract

DC arc fault occurring in a photovoltaic (PV) system may cause fire accidents, which should be detected and isolated timely. A novel DC arc fault detection method using Pseudo Wigner-Ville distributed (PWVD) algorithm is proposed. Arc fault experiment platform in a 5 kW grid connected PV system is built, and the circuit current of the PV strings under different test conditions are measured. The PWVD algorithm is used to analyze the currents. The time frequency graph of the current can be obtained, and the characteristic parameters for arc fault detection are extracted. The PWVD results can intuitively reflect the time, frequency and amplitude information of the current. The proposed method can detect DC arc fault in PV system accurately.