Short Circuit Fault Location Method of Distribution Network Based on Fuzzy Matching

Facing the difficult problem of short-circuit fault diagnosis and imperfect information environment in 10kV distribution network, a short-circuit fault section location method based on fuzzy matching is established by using the multi-source data of distribution and consumption information system of distribution network and taking Elman neural network model as the center. Firstly, based on the database of power distribution information system, the feature database of distribution network short-circuit fault diagnosis is established, and then the main features are selected by i-relief algorithm as the data input of Elman neural network. Finally, the multi-source data and the type and location of distribution network short-circuit fault are fuzzy matched based on Elman neural network model. Finally, through the analysis of a practical example in Southwest China, it is concluded that the proposed model can efficiently and quickly locate the short-circuit fault of 10kV distribution network, and has good fault tolerance and practicality.