Short-term passenger flow forecast of urban rail transit based on BP neural network

This paper discussed the framework of forecast model of BP neural network, represented a method for short-term passenger flow forecast of urban rail transit based on this model, which used the Matlab Neural Network Toolbox, took the influence of weather, date and other factors into account to forecast the passenger flow of Beijing urban rail transit. Through the experiment, we analyzed the results of the training time and the absolute mean-square error of using number of different neurons in the hidden layer, and analyzed the results of the fit, etc. of using different training epochs and learning rate, in order to decide the optimal number of neurons in the hidden layer, the training epochs and the learning rate. The good consistency between the forecast result and the actual value has been obtained. Therefore, the model can be used to forecast the short-term passenger flow of urban rail transit.