Fault Line Selection Based on Feature Fusion and Clustering
Linjing Hu, Wen‐Chen Hu, Hongyu Chen · 2024
Phase-to-ground faults are the most common in indirectly grounding power system. In order to protect the line and maintain the stability of power supply, it is very important to respond quickly after fault. In this paper, the zero-sequence current of single-phase-to-ground fault is analyzed and decomposed by CEEMDAN. The transient energy of modal components is calculated and the standard deviation of effective components is calculated by correlation analysis. Finally, The K-means clustering algorithm was used to classify the errors. Tests have shown that the algorithm can accurately select fault paths by considering external conditions such as fault starting phase angle, transient impedance, and fault location.