【作者】 黄荣；

【导师】 毛保华；

【作者基本信息】 北京交通大学 ， 交通运输规划与管理， 2010， 博士

【摘要】 城市轨道交通系统在发展过程中,由单一线路逐步形成经纬交错、线路之间联系密切的网络形态,其适用的线路运营组织理论与方法、技术也更加复杂。目前,国内外相关研究领域的学者已经对多线换乘、共线运营、多交路运营、快慢车结合、可变编组、共用车辆段、票务清算等网络化运营的相关组织方法与实施技术进行了不同程度的探讨,然而总体上缺乏从微观(车站、线路层面)到宏观(线网层面)、从单一实施技术到综合运用的研究,以及综合运用多种方法与技术产生的通过能力、运营效益等方面的不同效果研究。有鉴于此,本文研究了城市轨道交通线网不同的几何形态对线网换乘能力的贡献差异,将单线路的客流特征分析综合拓展到线网覆盖的整个市域范围；在系统阐述各种网络化运营的相关方法与技术的基础上,提出了网络化运营环境下的票款清分算法及实施方案流程；研究了共线运营结合多交路的两线通过能力损失计算方法,并提出了实现该运营环境下多交路线路通过能力最大化的判定条件和行车组织方法；研究了乘客总体时间效益最大化的快车跨站选择模型并给出了相应的算法。主要的研究工作及结论包括：(1)提出并分析了以两两线路之间的换乘结点数定义的线网换乘便捷性矩阵。针对线网形态中的网格平行线、放射线和环线和换乘车站中的两线换乘、三线换乘,分别探讨了其换乘便捷性的差异。计算结果表明,轨道交通线网中,增加不同线形的线路对线网换乘便捷性的边际贡献不同。以线网规模相近的东京营团地铁和北京城市轨道交通为对比案例,计算结果后者的换乘便捷性仅为前者的40%。分析表明造成这一差距的线网形态原因在于前者以网格型形态为基础,缺乏弧线与对角线线路,同时换乘结点以两线换乘为主。(2)从长大线路的不同方向的客流空间分布特征出发,研究了线网的总体客流空间分布特征。研究了连接郊区的长大线路的客流特征,其全线客流空间分布不均衡；通勤客流对早晚高峰的影响尤为突出；各区段之间(特别是市区客流与郊区客流之间)的客流交换量明显不均衡。在此基础上,连接各线路的最高流量客流断面,形成围绕城市中心区的一个闭合环线或大弧度曲线。(3)研究了共线运营环境和换乘环境下的票款清分算法,分析了线路与经营核算实体的对应关系,对换乘环境下的票款清分算法进行了改进。在构造算法的过程中,提出了共线运营区间“虚拟路径”的概念,对应相同OD的不同经营核算实体。从而利用多路径选择概率模型或者随机用户平衡模型,确定客流分配,实现共线运营环境下的票款清分。设计算例验证了本文改进的票款清分算法的有效性。(4)以长短交路嵌套的多交路线路为研究对象,给出了各区间通过能力损失的计算模型；提出了通过延长长交路折返时间达到线路通过能力最大化的方法；推导了采用这一方法的充分条件。算例表明,通过延长折返时间110s,线路单双交路区间单位小时通过能力分别提高了0.4列和1.2列,验证了最大化方法的有效性。(5)以郊区线采用过轨运输结合多交路的市区线共线运营为研究对象,给出了各区间的通过能力损失计算式；提出了市区线利用郊区线的过轨交路达到全线线路通过能力最大化的方法；推导出了采用这一方法的充分条件。此时,市区线全线通过能力损失降低为零；郊区线的通过能力损失增大,但增值低于市区线的降值。算例表明,在损失郊区线通过能力0.3列／h的条件下,提高了市区线原单交路区间通过能力13.3列／h,验证了最大化方法的有效性。(6)从整条线路全体乘客的角度,构建了实现旅客总体时间效益最大化的快车跨站选择模型。利用多目标规划问题的特点,采用将其转化为一个单目标问题的方法设计算法。跨站模型同时还可用于判定线路是否适用快慢车结合方法。设计算例,确定快车跨站方案并仿真铺画运行图,验证了本文模型及算法的有效性。更多还原

【Abstract】 In the construction process of urban rail transit system, a network with strong relevance among lines gradually comes into being from only a few lines having poor relation, and the applicable organization modes, implementation methods and technologies of network operations become more complex.The organization modes and implementation technologies of urban rail transit network operation, such as multi-line transfer, joint operation, multi-routing operation, express/slow train operation, train reformation, depot sharing, income assignment, have been discussed in varying degrees, some of which have been deeply studied. However, there are few studies on the comprehensive application of these methods and the corresponding effects like carrying capacity and operating efficiency. Therefore, contributions of several network geometric forms to transfer capacity as well as the passenger flow characteristics of urban rail transit network expanding to suburb were discussed firstly. Secondly, income assignment algorithm and its implementation plan process under the network operating environment were put forward according to the methods and technologies of network operation. Thirdly, the carrying capacity loss calculation method of two lines with joint operation combined with multi-routing operation was studied, and the approach to achieve the maximum carrying capacity and its conditions were brought forward. Finally, a model of express train stop station selection to minimize total travel time and corresponding algorithm were also studied.The main studies and conclusions are as follow.(1) Defined network accessibility matrix by transfer nodes between each two lines. The different accessibilities of parallel, radiation and ring lines in network, as well as two-line transfer and three-wire transfer in transfer station were explored. The results show the different marginal contributions of varying lines to accessibility in rail transit network. Take Tokyo Metro and Beijing urban rail transit networks for example. These two networks have similar sizes, while by comparison, the result shows that the accessibility of the latter is only 40% that of the former. The reason for this difference was analyzed, and relevant recommendations were put forward.(2) The Passenger Flow Distribution (PFD) characteristic of network is comprehensive manifestation of PED of lines in different directions. The study on PED of long lines expanding to suburb shows that the PED of whole line is unbalanced, commuter flow has special impact on the peak flow in the morning and evening, and passenger exchange volumes are obviously uneven among different sections, especially between the urban and suburban sections.(3) According to the essential difference of income assignment under the transfer environment and joint operation environment, an improved income assignment algorithm was proposed. In the course of algorithm construction, a "virtual path" concept was brought forward, corresponding to different operation accounting entities of the same OD. Thus passenger flow distribution could be determined by multi-path selection probability model or stochastic user equilibrium model to achieve income assignment under joint operation environment. Case study shows the validity of the improved algorithm.(4) The carrying capacity of multi-routing line was studied, and the calculation formulas of carrying capacity loss both in single routing interval and double routing interval were put forward. After that, a method to achieve the maximum carrying capacity by extension of back-turning time of long routing was brought forward and its sufficient conditions were deduced. Case study result shows that by extending back-turning time to 110s, carrying capacity loss of single routing interval and double routing interval is reduced by 0.4 trains per hour and 1.2 trains per hour respectively, which verify the validity of the method.(5) Suburban rail transit joint operated with urban rail transit of multi-routing was studied and the calculation formula of carrying capacity loss in each interval was put forward. Then, a method to achieve the maximum carrying capacity by over-rail routing of suburban rail transit line was brought forward, and its sufficient conditions were deduced. Thus carrying capacity loss of urban rail transit line was decreased to zero while that of suburban rail transit line was increased, however, the increased value was lower than the decreased one. Case study result shows that the carrying capacity loss of single routing interval of urban rail transit line is reduced by 13.3 trains per hour by increasing 0.3 trains per hour of carrying capacity loss of suburban rail transit line, which verify the validity of the method.(6) Travel time of the travelers all over a rail line was studied and a model of express train station selection to minimize total travel time was built. This model about multi-objective programming problem was transformed into a single-objective one to design algorithm. This model can also be used to determine whether express/slow train operation is applicable for an urban rail transit line. Case study verifies the validity of the model and algorithm.更多还原