Modeling And Empirical Research On Passenger Flow Distribution In Metro Network Based On Data Drive

With the construction of rail transit networks in various cities,the scale of its road network has gradually expanded,and transportation network operation has become a development trend.The shift from single-line operation to networked urban rail transit has caused multiple paths between the starting and ending stations,the road network has gradually become more complicated,and there are a variety of travel paths for passengers to choose from.However,the diversity of passenger paths makes the path flow difficult to count.Higher requirements are placed on the operation and management of urban rail transit.Therefore,the study ofpassengers’ route selection rules and the distribution and prediction of traffic flow in urban rail transit networks provide strong support for mastering passenger flow characteristics and road network passenger flow operation management.However,there are few studies on dynamic passenger flow allocation in subways,which is the motivation of this paper.First,introduce the AFC(Automatic Fare Collection System)data,timetable data and road network data of urban rail transit.For the problems in the original data,preprocess and analyze the original data to obtain the OD traffic volume by time period,which is the passenger flow Distribution lays the foundation.Secondly,the urban subway network structure is relatively complicated.In order to search for routes efficiently,the road network structure of urban rail transit needs to be simplified.Based on the shortest route search algorithm and effective route search algorithm,a space-time route search algorithm is proposed,and the case analysis and verification are carried out.Thirdly,the Logit model is used to calculate the dynamic passenger flow of the urban rail transit network,the general cost of the space-time path is constructed considering the factors ofthe travel time,transfer time,waiting time and in-and-out time,and the choice probability of the space-time path is obtained using the Logit model.Furthermore,the flow and method of dynamic passenger flow allocation in the subway network are proposed.Finally,the Beijing subway network is taken as the research object,and the algorithm is empirically analyzed.The distribution results obtained are compared with the data of Beijing Rail Transit Command Center to calculate the error rate,thereby verifying the accuracy and effectiveness of the method.