Passenger Flow Forecast of Urban Rail Transit Based on Wavelet Neural Network

As the basis for ensuring operational safety and improving management level, passenger flow forecasting is important in the organization of urban rail transit. Therefore, this paper studies a prediction model and wavelet neural network (WNN). Firstly, based on wavelet neural network, a prediction model is constructed and a prediction solution process is designed. Secondly, the number of nodes in input, output and hidden layer is tested and determined. Finally, taking Nanning urban rail transit as an example, combined with Matlab software, historical data of passenger flow is input to predict the daily passenger flow of a certain line, and then the forecast result is obtained and the error is analyzed. The result shows that the wavelet neural network forecast method improves the prediction accuracy, and the overall prediction error is lower than 10%. The accuracy of the prediction model is high, which can meet the demands of the operation management department to understand the distribution of the passenger flow with precision.