考虑跨站越行和同步协调的城市轨道交通列车开行方案建模研究

发布时间：2018-05-18 07:21

本文选题：城市轨道交通 + 建模　；参考：《北京交通大学》2017年博士论文

【摘要】：城市交通拥挤问题突出表现在影响人们日常出行质量和延误刚性出行效率方面,同时它还会诱发交通安全和环境污染等社会问题。城市轨道交通以其大运量、快捷、安全、高效、节能和环保等优势,已成为缓解城市交通压力的主要途径,并一直发挥着城市交通的骨架作用。随着城市轨道交通网络化建设的不断发展,运营管理的策略、措施和方法也已从独立线路运营进入了网络化运营阶段,传统、单一的列车开行方案已越来越难以满足规模大、复杂性强的网络化客运需求。在当今城市轨道交通列车开行方案单一化和客流条件复杂化的趋势下,准确、高效和可量化评价的列车开行方案模型亟需建立。建模研究将有望成为运营管理部门预测、计划和编制列车开行方案的理论方法,为既有的列车运行图调整和优化提供技术指导和评价手段,以提高城市轨道交通的乘客服务水平,旨在实现城市交通和乘客出行的和谐发展。本论文基于城市轨道交通开行方案和列车运行图的编制原则、理论和方法,分析了"线网-列车-客流"的交互影响机理,研究了列车运行过程中的停站选择问题、到发接驳约束、冗余时间计划和安全追踪间隔等,总结了包括停站时间、换乘客流的走行行为和到达规律、冗余时间等关键指标在内的列车开行方案建模方法;从满足客流需求和保障运营效率的角度出发,构建了多场景的考虑乘客出行成本和企业运营成本的模型,为停站选择设计、无缝衔接的换乘和鲁棒性优化提供了量化评价指标和调整方法的支持。论文首先就运营管理阶段的开行方案、列车运行图和鲁棒性研究及应用评价进行了详细评述,分析了以列车发车频率、交路计划、停站方案、乘客满意度、等车时间、换乘效率和鲁棒性为目标的既有研究成果的建模思想和求解算法,较全面地总结了城市轨道交通列车开行方案研究的基本内容和方法论。本论文以跨站越行模式、换乘的同步协调问题和列车运行图鲁棒性为研究核心,据此提出构建列车开行方案模型的主要目标和研究方向,展开了较深入的分析。其次研究了独立线路条件下,跨站越行模式的城市轨道交通列车开行方案的优化问题,准确的计算了乘客的等车时间、在途时间、列车的周期运行时间等指标,以行车组织的到发约束条件为基础,将乘客出行成本和企业运营成本考虑到模型中,并以两者最小化为目标,提出了带有0/1决策变量的双目标非线性混合整数规划模型。采用理想点法,将双目标函数转化为约束条件不变的单目标函数,通过禁忌算法实现了大规模求解问题的计算。基于4种典型的客流分布场景验证了模型的可用性。对比分析了弹性跨站越行模式和固定跨站越行模式的列车开行方案运营效果。根据国外城市轨道交通某运营线的实际运营场景,选取了多种列车产品组合的开行方案,基于给定的客流需求,对比了站站停列车、快车和大站列车三种列车产品运营效益;比较了由上述三种列车产品组合的五种跨站方案方案的优劣;基于5种典型的客流分布的案例,分析了不同客流需求对方案的适用条件。提出了简明地通过确定首班列车出发时刻,以实现网络条件下列车同步协调成功次数最大化为目标的模型。并设计了该模型的遗传算法,通过优化前后的比较和基于CPLEX软件求解的对比分析,证明了改进的遗传算法具有较好的求解质量和较高的求解效率。再次提出了一种基于对换乘客流准确划分的城市轨道交通同步协调的优化模型,从客流需求角度以满足客流无缝换乘的目标。针对不同的换乘站,基于正态分布的换乘客流的到达规律,以总换乘等车时间最少为目标,将列车行车约束、客流划分和等车时间计算方法等作为主要输入条件,构建了带有0-1决策变量的非线性规划模型。运用改进的遗传算法与计算机模拟结合的方法有效解决了模型求解的问题。最后建立了一种考虑客流需求扰动的城市轨道交通列车运行图鲁棒性优化模型,基于等待乘客和列车的交互关系,计算列车的实际停站时间,通过对比计划停站时间,以判断列车是否晚点。若晚点,则通过所提优化模型调整列车在站间运行的缓冲时间,以实现提高列车运行图鲁棒性、同时避免额外增加运行图周期的目标。以城市轨道交通列车运行图扰动时间之和最小化为目标,构建了非线性混合整数规划模型,通过改进的遗传算法对问题进行求解。
[Abstract]:Urban traffic congestion is a major problem that affects people's daily travel quality and delays rigid travel efficiency. At the same time, it also induces social problems, such as traffic safety and environmental pollution. Urban rail transit has become the main way to alleviate urban traffic pressure with its large capacity, fast, safe, efficient, efficient, energy saving and environmental protection. With the continuous development of urban traffic network, with the continuous development of urban rail transit network construction, the strategy, measures and methods of operation management have also entered the network operation stage from the independent line operation. The traditional, single train operation scheme has become more and more difficult to meet the large and complex network passenger demand. Under the trend of the simplification of the train running scheme and the complex passenger flow condition, the train operation model of accurate, efficient and quantifiable evaluation needs to be set up urgently. The modeling research will be expected to be a theoretical method for the operation management department to predict, plan and compile the train operation plan, and to adjust the existing train operation map and to adjust the train operation chart. In order to improve the passenger service level of urban rail transit and to realize the harmonious development of urban traffic and passenger travel, this paper is based on the principle, theory and method of the planning of urban rail transit and train operation diagram, and analyzes the interaction mechanism of "line network train passenger flow", and studies the mechanism of the interaction mechanism of "line network train passenger flow". The problem of stopping station selection in the course of the train operation, to the connection constraint, the redundant time plan and the safety tracking interval, and to summarize the modeling method of the train opening scheme, including the stopping time, the travel behavior and the arrival rule of the passenger flow, the redundancy time and other key indexes, from the angle of meeting the passenger flow demand and guaranteeing the operation efficiency. In this paper, a multi scene model considering passenger travel cost and enterprise operation cost is constructed, which provides the support of quantitative evaluation index and adjustment method for stop selection design, seamless transfer and robustness optimization. First, the thesis is on the opening plan of the operation management stage, train operation diagram and robustness research and Application evaluation. In detail, the basic contents and methods of the research on the urban rail transit train case are summarized in this paper. The basic contents and methods of the study of the urban rail transit train case are summarized in this paper. The core of the study is the synchronization coordination and the robustness of the train operation diagram. According to this, the main objectives and the research direction of the train planning model are put forward, and the further analysis is carried out. Secondly, the optimization problem of the train opening scheme for the urban rail transit is studied under the independent line condition. By calculating the passenger's waiting time, the time of the train, the time of the train and the time of the train, the passenger travel cost and the operation cost are considered in the model, and the two objective nonlinear mixed integer programming model with 0/1 decision variables is proposed. The ideal point method is used to transform the dual objective function into a single objective function with constant constraints, and the computation of large-scale solving problem is realized by tabu algorithm. Based on 4 typical passenger flow distribution scenes, the availability of the model is verified. The operation of the train operation scheme for the crossing mode of the elastic cross station and the fixed cross station crossing mode is compared and analyzed. According to the actual operation scene of a foreign urban rail transit operation line, the opening plan of various kinds of train product combinations is selected. Based on the given passenger flow demand, the operation benefits of three kinds of train products are compared between station stop train, express train and large station train, and five cross station scheme schemes of the three kinds of train product combination are compared. Good and bad; based on 5 typical passenger flow distribution cases, the application conditions of different passenger flow requirements are analyzed. A model is proposed to realize the maximum number of successful times of the train synchronization under the network condition by determining the departure time of the first train, and the genetic algorithm of the model is designed, and the ratio is optimized before and after. Compared with the comparison analysis based on CPLEX software, it is proved that the improved genetic algorithm has better solution quality and higher solving efficiency. A new optimization model of urban rail transit synchronization coordination based on the accurate division of passenger flow is proposed, and the target of seamless transfer of passenger flow is satisfied from the angle of passenger flow. The same transfer station, based on the regular distribution of the transfer passenger flow, takes the least time as the goal of the total transfer. The train constraints, the passenger flow division and the calculation method of the waiting time are taken as the main input conditions, and a nonlinear programming model with 0-1 decision variables is constructed. The combined method effectively solves the problem of model solving. Finally, a robust optimization model of urban rail transit train diagram is established to consider passenger flow demand disturbance. Based on the interaction between the passengers and trains, the actual stop time of the train is calculated and the delay time of the train is judged by comparing the planned stop time to determine whether the train is late. If it is delayed, The optimized model is used to adjust the buffer time of train running between stations so as to improve the robustness of the train diagram and avoid the extra increase of the operation cycle. A non linear mixed integer programming model is constructed with the objective of minimizing the time sum of the disturbance time of the urban rail transit train diagram, and the improved genetic algorithm is adopted. The method is used to solve the problem.
【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：U292

【参考文献】