Predicting the Time Required to Pass Congested Road Based on Neural Network Algorithm

Abstract

In order to predict the duration of traffic congestion, this paper established a traffic congestion evaluation model based on cumulative ratio Logistic regression and a traffic congestion time prediction model based on BP neural network. Combining Pearson test, numerical combination, standard deviation method and other methods to solve the problem. Based on the measured data of Jinshui Road in Zhengzhou, the average error is 0.019m/ s and the prediction error rate is 0.15%, both within a reasonable range. The model can improve the accuracy of congestion time prediction and provide some help to real life.