High-Impedance Fault Detection Based on Nonlinear Voltage–Current Characteristic Profile Identification

High-impedance fault detection (HIFD) is crucial in an effectively grounded distribution system because of the potential threat of fire and electric shock. HIFD has been extensively researched for more than 30 years, and a harmonic component in zero-sequence current is typically used in detection algorithms since waveform distortion is often produced by arc flash. For a stable HIF, it would be invalid or has lower sensitivity because of the limited harmonic content in zero-sequence current waveform. To overcome this problem, in this paper, a novel detection algorithm is proposed for detecting waveform distortion in the time domain. First, for the analysis of arc flash, the solid dielectric electrical breakdown theory is verified to be more suitable than theories based on the heat accumulation theory, and a nonlinear arc model that can be used for analyzing arc flash is then proposed. Furthermore, a novel HIFD algorithm based on the identification of nonlinear voltage-current characteristic profiles (VCCPs) is proposed, with the focus on the quenching and restrike dynamic process of an arc flash. The high performance of the algorithm is verified by performing various simulations and examining field data. An HIFD prototype using the proposed algorithm was developed, and it showed excellent performance in real-time digital simulator tests. The VCCP based disturbance detection approach could also be popularized in the applications of relay protection, harmonic source location and early fault alarming in smart grid.