| Along with the accelerating urbanization process,urban public transport network,as one of the important urban transportation network is faced up with the rapid growth of the dynamic traffic and imbalance of the time-space distribution of the passenger flows.As a result of the limitation of time and space resources,the existing public transit networks are difficult to meet the dynamic demands of passenger flows.On the other hand,as the variation of the topology structure of bus network,the density of transit network,the scale of transit network,the bus routes and stations,the bus schedule etc.,public transit network is becoming more and more complex,which require more efficient operation scheduling and robust network topology.Therefore,based on the dynamic passenger flow demands,optimizing robustness of urban public transport network with better planning and design and optimizing departure intervals of urban public transport network with better operation scheduling are extremely beneficial for urban public transport network operation efficiency,reliability,service level and improving the conditions of urban public transportation.Dynamic passenger flow demands are most significant data basement for above studies.However,it is also not easy to collect accurate dynamic passenger flow demands because of the structural data lackness like less effective travel data and lackness of accurate transfer passenger flows.Based on the studies of uran transit network and complex network theory by domestic and exotic researchers,with the basis of urabn transit network data such as Zhuhai in China(including urban road network topological structure,geographic information system data of transit routes,transit routes and bus station relevant data,up and down data of IC cards,global positioning system station arriving data etc.),this paper built up the urban public transit network with complex network theory.The model of generating OD matrix of urban transit passenger flows based on travelling chains is proposed and the model of generating transit transfer passenger flows based on data fusion,trajectory analysis and coefficient partition is put forward to derive traval eatimation of OD and transfer data of urban transit network in representative days.As a result,the structural data lackness was resolved and accurate dynamic passenger flow demands were derived.On this basis,this paper proposed the analysis index system of passenger flow characteristics for transit planning and operation decision in three aspects of passenger flow,bus operation,bus passenger travel characteristics.At the same time,the topological characteristics of the public transport network in Zhuhai are analyzed.Furthermore,this paper proposed a passenger flow loading model and algorithm of urban transit network driven by passenger flows to deal with the key issues like interval proportion partition of passenger flows,passenger flow distribution of the multiple-track routes,passenger flow distribution of different buses in the same route,calculation of up and down passenger flows and passengers on buses between adjacent two stations.Moreover,based on robust indexs,static robustness structural analysis and optimization under different attack modes was undertaken.Then under different attacking modes,dynamic robustness of urban transit network based on passenger flow loading(ROM-PFL)was studied and optimization advices was considered with structural analysis of topological robustness through dynamic passenger flow OD loading.Based on the optimization of cost,an urban transit departure frequency optimization model and particle swarm optimization algorithm under realistic elastic OD demands and capacity limitation(UTDFOM-REODD-CL)which optimized departure frequency of different bus routes in different intervals in representative days.The case studies in chapters of this paper show that the derived OTD with dynamic passenger flow demands in this paper built up a significant foundation for realizing the passenger flow loading model of urban transit network driven by passenger flows.ROM-PFL optimizes dynamic robustness of transit network in Zhuhai.Finally,UTDFOM-REODD-CL can realize the optimization of departure frequency of different bus routes in different intervals to improve the operation efficiency,cut down economic cost,satisfy dynamic travelling demands of pessengers and minimize travel time of pessengers.The main innovations of this paper are as follows:(1)The passenger flow OD matrix generation model based on trip chain and passenger flow OD matrix generation model based on data fusion,trajectory analysis and coefficient division was proposed to get the full OD and transfer passenger flows which can resolve the structural data lackness like less effective travel data and lackness of accurate transfer passenger flows.(2)Based on the comprehensive study of existing statistical eigenvalues,the topological structure of urban public transit complex network was calculated comprehensively from each existing statistical eigenvalues and the failure regularity with each statistical eigenvalues is also analyzed.(3)The urban public transit network loading model and algorithm were constructed driven by passenger flows.It could resolve some key items like time series analysis of urban public transit network and loading dynamic passenger flows,formulating departure intervals and transit schedule,computing time and space series of passenger flows,the coordination and operation of different transit routes and buses,calculating up and down passenger flows and amount of boarding pessengers of each bus.(4)Analysis of the static robustness and the variation of static robustness under different modes of attack was done.Through loading dynamic passenger flows,dynamic robustness variation of urban public transit network was analyzed under different attack modes.Furthermore,the advices for improving the topology of urban public transit network based on the robustness analysis was put forward.(5)Considering the dynamic demands,the departure intervals optimization model based on objective of cost of and particle swarm algorithm were proposed to realize the optimization of transit departure intervals with constraints of dynamic demands in different time intervals and different transit routs.Under constraints of transport capacity. |
PDF Full Text Request