Abstract：The identification of faults in distribution lines relies solely on a single current waveform analysis, which fails to fully consider the complexity of faults, and the identification time for global fault points is relatively long. Therefore, this paper proposes an automatic identification technology of faults for 10 kV distribution lines based on non-dominated sorting genetic algorithm Ⅱ (NSGA-Ⅱ). By analyzing the differences in current waveforms and using the method of comparing the correlation between current waveforms, the fault section in the 10 kV distribution line is preliminarily located. NSGA-Ⅱ is used to iteratively optimize the preliminarily located fault section and output the optimal solution for the preliminarily located fault section. Based on the optimal solution of the preliminarily located fault section and the dynamic characteristics of current and voltage after the fault, the electrical quantity such as starting current is calculated, and the electrical quantity information of each fault section is optimized and adjusted. Under normal and fault conditions, based on the set value, the fault point location is automatically identified through the singularity evaluation method using voltage and current changes. The simulated experimental results show that the proposed method can successfully detect faults in 10 kV distribution lines and complete the identification of all fault points within 3 seconds, which has significant advantages.
2025, Vol. 26 
