Research on Transient Feature Recognition Algorithm for Small Current Ground Fault in Distribution Terminals

Fault line selection and location are challenging in distribution networks with non-effectively grounded neutral points due to the extremely weak fault current during single-phase-to-ground faults. This paper proposes a method utilizing the time series singularity detection principle to identify single-phase grounding faults. The method accurately determines whether a single-phase grounding fault has occurred on a distribution line based on the magnitude and polarity characteristics of the zero-sequence voltage and zero-sequence current fault components. Real-Time Digital Simulation (RTDS) and dynamic simulation experiments validate the algorithm. Results demonstrate that the algorithm can correctly identify line grounding faults by detecting abrupt change points in the zero-sequence voltage signal and combining them with the polarity direction characteristics of the zero-sequence voltage and current fault components. The experimental results confirm the effectiveness of the algorithm.