Research on accident prediction of intersection and identification method of prominent accident form based on back propagation neural network

This paper introduces an intersection accident prediction model by applying BP neural network, which can well structure the corresponding function of intersection traffic conditions and accident form, and the paper trains and predicts the model with 197 intersection data, the test results show that the prediction accuracy can reach up to 89 percent. The paper establishes an identification method of accident-prone form according to the protrusion theory on the basis of the prediction result of accident number of sub accident form, which can provide basis for intersection adaption reconstruction.