Research on Identification of Single-Phase Grounding Fault in Distribution Network

Single-phase grounding is one of the most common faults in power distribution system, and it is of great significance to identify single-phase grounding accurately and quickly. However, there are several types of disturbances in distribution systems that can produce fault characteristics similar to single-phase grounding. Aiming at several types of disturbances that may cause misjudgment of single-phase grounding, this paper proposes a disturbance identification algorithm based on prior knowledge. Firstly, the effective expansion of various disturbance samples is realized based on the probability distribution of disturbance prior knowledge. Secondly, a decision tree screening link is established based on prior knowledge to filter out untraceable disturbances. Finally, the multi-kernel SVM algorithm is used to identify single-phase grounding and its interference. The simulation results show that the proposed identification method can overcome the influence of unbalanced samples and unknown disturbances on the identification accuracy, and the algorithm has good stability.