Abstract:

Addressing the challenges posed by the subtle characteristics of single-phase grounding faults in resonant grounding systems, which lead to decreased sensitivity and reliability in fault line selection methods reliant on these characteristics, an innovative approach for fault line selection through active injection signal identification is proposed. Initially, the distribution characteristics of the injected current, influenced by line-to-ground conductance, are derived. These characteristics then form the foundation for constructing a robust line selection criterion. Furthermore, a thorough analysis is conducted to assess the impact of the injected signal on system operation from various perspectives. This analysis leads to the selection of an optimal injection signal with suitable parameters. Notably, to enhance the method's resilience against transition resistance, a decision is made to inject a low-frequency signal. Subsequently, the zero-sequence current of each feeder line is measured after injecting the signal. A specific steady-state zero-sequence current time window is identified, and the Prony algorithm is employed to accurately discern the 25Hz current amplitude within the zero-sequence current of each line. This approach enables precise fault line selection. Extensive PSCAD simulations and on-site waveform measurements validate the effectiveness of this method. The results demonstrate its ability to accurately identify fault lines across a range of fault scenarios, while maintaining excellent resistance to transition resistance.

Key words: distribution network, single phase ground fault, injection method, Prony algorithm identification, fault line selection