【作者】 漆凯；

【导师】 张星臣；

【作者基本信息】 北京交通大学 ， 交通运输规划与管理， 2012， 博士

【摘要】 我国经济的飞速发展与社会交往活动的日益频繁极大地刺激了交通运输需求的增长,综合交通系统也已进入了一个快速发展时期。在发展与完善区域性综合交通运输网络的过程中,多种交通方式汇集的综合客运枢纽建设与管理是一个不可缺少的组成部分。作为联系多种交通运输方式的纽带,城市客运枢纽站需安全高效地提供旅客乘降服务,并发挥大客流时的旅客集散与中转功能。因此,在网络化运营的背景下,“快在中间,慢在两端”的实际情况对城市客运枢纽站的一体化运营提出了更高的要求。综上,在完善基础设施网络的同时,系统地研究客运枢纽站旅客流线优化技术与方法,制定合理高效的旅客流线方案,将有助于改善枢纽站的客流组织,发挥枢纽的集散与中转功能,提升综合交通系统运行效率及服务质量。本文以城市客运枢纽站旅客流线为研究对象,重点研究了枢纽站旅客流线优化的相关技术和方法。论文的主要工作包括以下几方面：(1)在既有枢纽分级标准研究的基础上,阐述了枢纽等级划分的基本原则,提出了以功能定位、能力、占地为评价指标的城市综合客运枢纽分类与分级标准。在此基础上,探讨了枢纽内流线优化同枢纽等级之间的关系,分析指出了对于不同等级的客运枢纽,流线优化的难点、复杂程度存在差异；并研究了枢纽内设施布局与旅客流线优化的关系,指出枢纽建筑空间布局、功能区划分及交通吸引发生源的位置是客运枢纽旅客流线优化的基础。(2)在分析枢纽站旅客流线广义费用影响因素的基础上,将枢纽路径广义费用分为路段和节点两部分。在BPR路阻函数的基础上,考虑了旅客在通道中对拥挤和服务水平的心理感知因素,构建了路段广义费用函数；针对枢纽节点,将其抽象为有流量无空间尺寸的虚拟点,构建了基于Webster模型的节点广义费用函数。参数分析表明：枢纽站路径广义费用值随着拥挤度和上坡坡度的增大而增加,且变化速度呈现几何增长趋势,表明路径广义费用函数在反映旅客走行时间的同时,也反映了心理因素对旅客出行中路径选择的影响。(3)枢纽站内旅客流线优化问题的核心是怎样在整个系统的角度上设计旅客流线,使得整体系统的总费用最小。提出将城市交通流分配模型应用于枢纽站旅客流线优化的研究中。在对比与城市道路交通流分配差异的基础上,指出枢纽站旅客流线优化实际为单路径容量限制的系统最优流量分配模型；在此基础上,构造了基于流量分配的旅客流线优化模型,最小化系统广义费用作为优化目标,采用K条渐短路算法确定备选路径集合Krs,,旅客流量按照单路径分配,确定旅客最优流线优化,并通过具体案例对模型进行了分析。通过对模型中总体OD需求和单OD需求的不确定分析表明：总体OD需求增加时,系统的总费用呈现指数型增长,节点广义费用在总费用中所占比例由16.61%增大到49.54%；单OD需求变化越大,单位流量的增加使得系统的总费用的增幅越大,且旅客在节点处的费用占整个网络总费用的比例显著增大,节点成为制约网络效率和流线优化的关键。(4)将由统计物理(Statistical Physics)发展而来的最大熵原理应用于枢纽站旅客流线的优化过程。在信息论熵方法的框架下,阐述了在相同广义费用约束下所对应流线方案的组合差异,以及由此导致的流线优化过程中信息熵产生的现象,并引入了旅客流线优化测度熵；在此基础上,根据最大熵原理,将旅客流线方案优化描述为最大化测度熵的优化过程,构建了旅客流线优化的最大熵模型,并给出了计算方法。通过对模型中OD需求和广义费用估计的不确定分析表明：流线优化过程中影响因素的信息不确定性决定了与枢纽建筑空间布局匹配程度高的方案将以较大的概率成为最优流线优化方案,而匹配程度是通过对OD需求和广义费用估计体现的。更多还原

【Abstract】 With the rapidly develop of China’s economic and the frequently of social interaction, the growth of transport demand is increased quickly. At present, China is in a period of diverse passenger transports with a quickly develop. Based on the improvement and improvement of regional integrated transport network, China has gradually built a number of city comprehensive collections of passenger hub with various transports. Contact a variety of modes of transport as a link, urban passenger transport hub to be provided safely and efficiently to passenger for take down service, and play transit and distribution function when there are a major of passengers. Thus, in the context of network operations, the actual situation of "fast in the middle and slow at both ends" put forward higher requirements on the integration of urban passenger hub operations. In summary, improving the infrastructure network, systematic studying the passenger terminal passenger flow routing design optimization techniques and methods and developing rational and efficient passenger flow routing program will help improving the organization of passenger flow, playing the distribution and transshipment function of passenger hub, and enhancing the efficiency and service quality of integrated transport system.In this paper, we study on urban passenger flow routing, focusing on the hub of passenger flow routing design and optimization of related technologies and methods. The main thesis works as follows:(1) Based on the study of grading standards of hub, the paper explain the basic classification principles, and put forward a classification and grading standard of urban passenger transport hub with evaluations (functional orientation, ability and area). On this basis, explored the relationship between the level and the flow line of hub, analysis the difficulty of flow-line design in the different levels of passenger hub; and studied the relationship between the passenger flow routing design and the layout of the facilities, then point out that architectural space layout, function zoning and traffic to attract the location of the source occurred of hub is the basic of passenger flow routing design in the hub.(2) Based on the analysis of generalized cost factors of passenger flow routing in the hub, the paper divide generalized cost into two parts (section and node). Considering the passengers’psychological perception factors of congestion and service levels, build a section generalized cost function with BPR resistance function; abstract the node as a non-size virtual point with flow, and build a node generalized cost function based on the Webster model. Parameter analysis showed that:with a geometric growth, the value of generalized cost increase when congestion and uphill slope increase. This point out that path generalized cost function not only reflects passenger travel time, but also reflects the impact of psychological factors in the path selection.(3) As the base of design and optimization, the distribution of urban traffic flow model is applied to the study of passenger flow distribution in the hub.Based on the comparative the difference of distribution between passenger flow and traffic flow, analyze that the nature of the passenger distribution is to consider single-path capacity limit of the flow distribution; then based on flow distribution, construct a optimization model of passenger flow, and determine the collection of alternative paths () using K gradually shortest path algorithm, the generalized cost minimization as the optimization goal, passenger traffic allocation by a single path, and then determine the optimal passenger flow routing design. The analysis the uncertainty of overall OD demand single-OD demand in the model shows that:the overall OD demand increases, the total cost of system showing exponential growth, the total cost of the node in the generalized cost increase form16.61%to49.54%; the more single-OD demand changes, the greater increase of total cost with the increase of unit, and the proportion of passengers’ cost at the nodes in the network total cost increased significantly, then nodes became the key faction to constraint the efficiency of network and optimize of passenger flow routing.(4) Maximum entropy principle, developed from statistical physics, was applied into the design and optimization process of passenger flow routing in the hub. Under the framework of entropy method of information theory, explain the differences of different flow lines combinations in the same generalized cost, and the phenomenon of entropy in the process of flow design optimization, then introduction the measure entropy of passenger flow routing design optimization; based on the maximum entropy principle, describe the passengers flow line design as a optimization process to maximize the measure entropy. Then, build a t maximum entropy model of passenger flow routing design optimization, and gives the calculation method. Through the uncertainty analysis of OD demand and general cost estimate of the model shows that:because of the uncertainty of the information of factors in the passenger flow routing design optimization process, hub architecture with a high degree of spatial layout of the matching program will have a greater probability to be choose as the optimal design, and the degree of matching is estimated by OD demand and generalized cost.更多还原