Faulty feeder detection of single phase-earth fault based on fuzzy measure fusion criterion for distribution networks

Abstract

Faulty feeder detection of single phase-earth fault for medium voltage distribution networks is still challenging since the weak fault feature and diverse fault conditions. A novel single phase-earth fault feeder detection method based on fuzzy measure fusion criterion is proposed in this paper. Various historical feature samples which characterize different fault conditions of the protected feeder are divided into fault group and non-fault group by fuzzy c-means clustering algorithm. The similarity between real time data and the historical feature sample set is quantitatively represented by the fuzzy measure fusion criterion. Fuzzy measure fusion criterion matrix is evaluated through a multi-level evaluation index system to emphasize the effective fault information and reduce the impact of accidental factors. The detection criterion is established by comparing the similarity between the detected feature sample and the historical feature samples to identify the faulty feeder of single phase-earth fault. PSCAD/EMTDC simulation and laboratory fault experiment results have confirmed the effectiveness and adaptability of the proposed detection method. The accurate fault detection can be realized even in the rigorous fault cases of high impedance grounding fault and arc grounding fault.