Fuzzy Inference Model for Recognition of Single-Phase-to-Ground Fault Caused by Insulation Deterioration in Medium-Voltage Distribution Networks

Abstract

Single-phase grounding faults seriously affect the power supply reliability of medium-voltage distribution networks. Insulation deterioration is one of the important causes of single-phase grounding faults. Based on the single-phase ground fault waveform data, eight features that can reflect the type of insulation deterioration are extracted from the time domain and frequency domain. The Receiver Operating Characteristic (ROC) is used to optimize the features, and the classification threshold of each feature is determined. A fuzzy inference system for identifying the causes of insulation deterioration in medium voltage distribution networks is established. The use of field-measured fault data verifies the necessity of feature selection and the validity of the established model.