The Study On Prediction Of Passenger Flow By RBF Neural Network On Shanghai Metro

China’s rail transit has developed rapidly in the era of rapid development of neural network algorithm,and has become one of the most important means to solve the problem of urban traffic.The residents travel more and more dependent on the rail transit,the service quality of service requirements are getting higher and higher,more and more operational management pressure.The rapid growth of passenger traffic makes the road network load increasing.The contradiction between transport and transportation will continue to be realized during the rapid growth of rail transit.Timely grasp the real-time passenger traffic distribution of rail traffic,passenger dynamic development can help the Shanghai rail transit management department or other relevant departments to do a good job of passenger organization,scheduling and emergency response.This requires a basis for passenger forecasting.In this paper,based on the short-term prediction of the traditional forecasting model,the accuracy of the short-term prediction is not high and the prediction scale is too large.The passenger flow forecasting model based on the radial basis function neural network is put forward.First,The mathematical model is used to make the radial basis function neural network model.The Gaussian function is used to calculate the radial basis function model.The non-supervised K-means clustering method is used to train the radial basis model Select the representative parameters as the center,with the trial and error method to determine the expansion constant,through the training of RBF model to get learning weights.Finally,the model is applied and verified by the Yishan Road Station on the 4th line,and the prediction accuracy of the model is improved by adjusting the number of the center,the number of layers hidden and the learning weights.By preconditioning the input data,The prediction speed and accuracy of the model are improved by adjusting the prediction scale and the prediction accuracy of 30 minutes is controlled within 5%.The results show that the model can effectively solve the problem that the accuracy of the passenger flow prediction is not high and the prediction scale is too large.