Study On Passenger Flow Modes And Evolutionary Mechanism Of Urban Rail Transit Networks

The space-time distribution of passenger flow is not only one key point of the high-efficiency and safety of urban rail transit operation,but also the basis of line operating plans.The system behavior and status evolution mechanism of passenger flow in station,line and network are analyzed through the distribution,and the corresponding control measures are provided,which are of theoretical significance to achieve the efficient co-operation among lines,improve the network transport efficiency and service quality,and guarantee operational safety.Therefore,based on urban rail transit network topology model,firstly this paper studies the passenger flow patterns,including station(microscopic),line(medium)and network(macroscopic)passenger flow modes.Secondly,the modeling of space tie-in attraction characteristics regarding passenger flow in urban rail transit network based on field theory is provided and the calculation method of modeling parameters is also given.The last but not the least,this paper presents two evolution models of network passenger flow modes in different operation scenarios,i.e.,normal operation scenario and train delayed operation scenario.Finally,modelling computation and verification are given based on the actual data.Specifically,the main contents are as follows:(1)Passenger flow modes and control strategies in urban rail transit stationThe main variables such as the station hall,passageway and platform passenger flow,are taken to analyze the passenger flow non-linear interaction and the passenger flow evolutionary model is established.The analytical solution of the model and stability condition of the model's equilibrium points are solved to figure out the mode set.Then the physical characteristics description of the modes is analyzed.And the mode transition threshold condition is given.The numerical simulation shows that when the passenger value of entering in station exceeded the threshold value A0,and the difference between passenger values of getting on and off the train is larger than ?C0,the passenger flow status develops towards the mode level III,which means a high load operation state.(2)Passenger flow modes of urban rail transit line based on gain-weighted comprehensive methodCombined with passenger flow of sections and stations,the topology of line,train operation and the coordinating passenger flow among lines,a double-layer line index system is proposed for line passenger flow modes.Regarding to the characteristics of index systems,Gain-weighted Comprehensive method is used for mode calculation.According to the threshold of each level factor,line passenger flow modes are divided into three levels,and the critical points of each mode level are determined.(3)The modeling of space tie-in attraction characteristics regarding passenger flow in the network of urban rail transport-stations based on field theoryAimed at the complicated problems of correlation attraction characteristics regarding passenger flow space among the networked operation stations,the concept of the gravity field of passenger flow is proposed for the first time to conduct a research into the internal characteristics and operation mechanism.We establish the computation methods of measure of characteristics of passenger flow(including field intensity and potential energy),with the aim of the quantization of the passenger flow potential of the stations from the perspective of the collection and distribution of passenger flow and the topology of network.As for the computation methods of field intensity,we have proposed an optimum route based on data verification of OD and cross-section passenger flow volume to define Betweenness Centrality parameter,and the model works best when ?=1.531.Regarding the computation of potential energy,we have applied Compound Simpson's Rule Formula to get a solution to the function.With random process methods,the expectation value has been used to conduct a quantization of the influence of passengers' getting off on the potential energy of the source point of passenger flow.Taking the subway line 10 of Beijing as an example,the paper validates the model.(4)Network passenger flow modes evolution mechanism of urban rail transit in a normal operation scenarioThe passenger flow exchange model under the condition of network operation is established.Given that it is hard to obtain actual data,based on the calculation formula of passenger flow field,this paper estimates the proportion of passengers getting off the train,which is in passenger flow exchange model.The dynamic evolution model of passenger flow based on cellular automata is constructed to study the dynamic propagation and evolution.Taking part of the urban rail transit network in Beijing as an example,we validate the physical characteristics of network passenger flow at different times by adjusting the parameters co of passenger flow getting on the train.(5)An evolution model of network passenger flow modes in a train delayed operation scenarioIn order to describe the overall passenger load status of the whole network,considering topology attributes of the nodes and edges and the flow throughput,the load calculation formulas are put forward,and a description method of the network passenger flow mode based on the load entropy measure is proposed.Considering the parameters such as the optimal path and the actual train capacity function of the line,the calculation method of the increment of the load is put forward,and the passenger flow mode evolution model of the urban rail transit network under the train delay operation is established to describe the transition between different modes.The results show that the simulation results obtained by the passenger flow mode evolution model are consistent with the actual passenger flow conditions at different times(7:30-9:25)in a train delayed operation scenario.Three typical mode transitions occur at 07:50 ?07:55,07:55 ? 08:00,08:55 ?09:00.The time of the transition,the response time after the failure occurs,and the time of measures being taken in different stages highly matches the mode transition time.