Prediction And Analysis Of Passenger Flow In Urban Rail Transit

With the development of social economy and the rapid advancement of urbanization,the urban population is increasing,making urban traffic problems more and more serious.Vigorously developing public transportation has received the support of the government.Urban rail transit has become the main direction of public transportation in major cities because of its lack of land resources,safety,pollution and strong transportation capacity,which has alleviated traffic congestion to a certain extent.However,with the increase of the scale of the urban rail transit network and the increase in passenger traffic,the transportation pressure of rail transit is also increasing.How to effectively improve operational efficiency,rationally allocate resources,timely adjust the transportation capacity of rail transit,and rationally arrange personnel,which has become a difficult problem to be solved.Based on the complete passenger flow data on AFC(railway self-service ticketing system),this paper analyzes the distribution law of passenger flow in Xi'an urban rail transit,establishes a time series forecasting model based on BP neural network,and predicts the short-term passenger flow.The work arrangement of the person and the passenger travel provide a reference basis.The main conclusions of this paper are as follows:(1)Based on the Xi,an rail transit passenger flow data obtained from the subway comPany,this paper summarizes the general passenger flow distribution characteristics of Xi'an urban rail transit,and then analyzes the passenger flow distribution characteristics of Xi,an Metro Jixiang Village site in one day and one week,and finds auspiciousness.The passenger traffic at the village station is relatively large,especially on working days,which are generally double peaks.(2)Introduce the commonly used methods of traditional prediction and micro prediction,as well as the applicable eonditions of each model and their advantages and disadvantages.It is found that the different prediction theories have different emphasis,so the scope of application and the accuracy of prediction are different.The passenger flow of rail transit has dynamic characteristics,and the combined forecasting model has an advantage,and the advantages of each single model are concentrated to better predict the passenger flow.(3)According to the above analysis,when the passenger flow prediction modeling is carried out,the neural network and time series are combined to establish a time series prediction model based on BP neural network,and the prediction results are obtained by using MATLAB software.The results show that the error between the predicted value and the actual value is 12.99%,which is more accurate than the traditional prediction model.