Reseach On Optimization Approaches Of Urban Rail Transit Traffic Organization Based On Passenger Flow Prediction

With the rapid development of urban rail transit and the continuous improvement of passenger service requirements,the requirements and organizational management level of urban rail transit operation resources are increasingly refined.It is necessary to find the balance between supply and demand of rail transit line resources and passenger flow changes,to effectively improve the quality of passenger service,and reducing operating energy consumption level.Therefore,the optimization of passenger flow organization is an important research direction.With the application of urban rail transit information technology,Taiyuan rail transit formally introduced the urban rail cloud big data platform when the first line was constructed.Since the line was operated,a large amount of data has accumulated,making it possible to analyze data based on actual passenger flow data.Indeed,mining and forecasting work becomes possible.This paper combines the passenger flow information collected by the big data platform for data extraction and analysis,and analyzes the temporal and spatial characteristics of actual passenger flow data from the time distribution of passenger flow at each station and the distribution of passenger flow in and out of the station at each time period.Moreover,the short-term passenger flow changes are accurately predicted based on the passenger flow prediction model,which provides reasonable adjustment basis and optimization measures for the actual train organization of the line,and solves the problems such as the mismatch between passenger flow and line resources and bad experience of passengers.The main research contents of this paper are as follows:(1)Based on the actual operation data,the distribution characteristics of urban rail transit passenger flow are deeply excavated,and the actual passenger flow of Taiyuan Rail Transit Line 2 in 2021 is used as the analysis object to analyze the change trend and passenger flow characteristics,the fluctuation of the passenger flow of the line in a week,and changes in passenger flow at typical stations with large passenger flow.(2)Combined with the nonlinear and non-stationary time series characteristics of the actual line passenger flow data,the advantage of ARIMA not based on other exogenous variables is used to avoid the failure to capture the nonlinear relationship,and avoid the shortcomings of the model such as low prediction accuracy caused by multi-order differences,a MM-ARIMA short-term passenger flow model based on the combination of the forecasting model and the Markov model is established.(3)On the basis of accurately predicting the change of passenger flow,combined with the prediction results,the adaptive countermeasures of traffic organization and passenger flow characteristics are studied,aiming at the optimal allocation of transport resources and the accurate matching of transport capacity and capacity,and propose the optimization scheme of traffic organization: the scheme of adding small traffic routes and flat traffic routes during peak hours;adjust the morning and evening peak times and train intervals;research on the dynamic adjustment of train marshalling forms in different peak sections under the demand of passenger flow,so as to realize the precise matching of line capacity and volume in the case of meeting passenger demand.