BP Neural Network Based Fault Prediction of Distribution Network during Typhoon Disaster

Typhoon disasters will cause faults such as reversing, disconnection, wind flashover, etc., which threaten the safe and reliable operation of the distribution network. It is necessary to predict the failure of the distribution network caused by typhoon disasters. Considering that a large number of historical data from typhoon and distribution network fault information are retained in the meteorological information acquisition system, it is very suitable to use artificial neural network to explore the correlation between meteorological information and distribution network fault information. Therefore, this paper proposes a fault prediction model of distribution tower during typhoon disaster based on BP neural network. The fault prediction model of distribution network based on BP neural network is established firstly. Then the parameters in the model are optimized by training samples. Lastly, the validity of model prediction is tested by test samples. The test results show that the proposed method can accurately predict the fault trip of the distribution network tower during typhoon disasters.