Research On Time And Space Distribution Of Passenger Flow Assignment Model In Rail Transit Network

China's population growth has led to an increasing contradiction between urban metro transport capacity and passenger travel demand,especially during peak working days.Station passenger congestion,channel passenger flow queuing,and passenger flow emergency control often occur.Understanding the current state of passenger flow distribution needs is the basis for solving these problems.With the increasingly complex network operation of rail transit,passengers may have multiple optional effective paths between the starting points of the orbital network,which makes it difficult for the urban rail passenger flow clearing work.At present,most domestic and foreign scholars' research on passenger flow data focuses on passenger flow forecasting,passenger flow distribution feature analysis and passenger flow simulation.The research scale stays at the total passenger flow characteristics of the station.Based on the above background,this paper innovatively deduces these three aspects from the analysis scope,verification method and distribution result level,and makes a more comprehensive and in-depth thinking than the previous passenger flow allocation research,and proposes a comprehensive algorithm framework.The research results can effectively conduct a comprehensive demonstration study on the distribution characteristics of passenger flow in Guangzhou rail transit.Combined with the information collected by passenger flow,time and space,the passenger flow intensity of urban rail transit interval,line and line network is comprehensively evaluated.The specific work is as follows:This paper mainly studies the passenger card data collected by the Urban Rail Transit Automatic Ticketing System(AFC),and finds the effective route passenger flow distribution ratio,in order to achieve the research purpose of clearing the passenger flow between the effective paths.Based on the excellent results of domestic and foreign scholars in the research methods of passenger flow distribution,passenger flow distribution characteristics and congestion identification,the topological traffic model of Guangzhou rail transit is constructed,and the effective path set is identified and optimized.List storage.Based on the hypothesis verification that the travel time of a single path obeys the lognormal distribution,a hybrid model consisting of multiple normal distributions is constructed.The EM(Expectation Maximization)algorithm is used to perform iterative estimation of the parameters,and the data is corrected using the 3? criterion.The improved finite Gaussian mixture model(LGMM)performs an example analysis of the data and verifies the parameter K value.In the application case analysis of the model,the field questionnaire data and the decision tree model based on individual selection are used to verify the accuracy of each algorithm.The model is extended on the full sample AFC data of the Guangzhou rail network to solve the current situation of passenger flow distribution.The calculation model of the passenger flow intensity index at the time and space level is constructed,and the passenger flow intensive traffic flow distribution characteristics of the urban rail transit network are comprehensively evaluated.The conclusions of this paper show the distribution characteristics of Guangzhou rail transit passenger flow in time and space.From the theoretical framework,the problem of passenger travel route selection in rail transit network is deeply discussed,which helps to accurately grasp the distribution law of passenger flow in rail transit network,set indicators and The analysis level has carried out comprehensive scientific research on passenger flow congestion,providing new ideas for daily passenger flow monitoring and guidance for risk assessment management in urban rail transit operations.