【作者】 陈晨；

【导师】 邓连波； 朱小山；

【作者基本信息】 中南大学 ， 工程（专业学位）， 2023， 硕士

【摘要】 随着公共交通不断发展,协同轨道交通与常规公交的运营是公共交通一体化发展的关键,可最大限度发挥二者优势,提高公共交通覆盖范围和服务质量,增强公共交通的吸引力。本文以客流多对多需求下的FBNDP问题为基础,提出带时间窗约束的城轨关联公交接驳线网(FBNDP-TW)优化问题。本文的主要研究内容如下:(1)为保障接驳公交网络的服务质量,对网络中客流OD的全程旅行时间进行严格限制。引入时间窗参数阈值δ,构造基于OD理想旅行时间的时间窗函数,逐OD设置差异化的时间窗约束。在经典接驳公交网络设计问题中,增加出行时间窗约束,构造了以乘客广义出行费用和公交运营商成本最小为目标的城轨关联公交接驳线网优化模型,优化确定接驳线路的接驳车站、线路经由、开行频率。(2)基于遗传算法框架,设计了带时间窗的城轨关联公交接驳线网的优化算法。为保障线路服务质量并适应时间窗的要求,设计了由直连、前向、邻接三种规则组成的多样性邻域交叉操作。区分时间窗约束是否满足的情形,选择不同的交叉规则组合策略,有效引导子代个体的进化方向,使网络中各点尽快满足时间窗要求,高效求解该问题。(3)以接驳线网标准算例验证了本文提出的优化模型和算法。结果表明,时间窗约束显著降低了各OD的绕行比,每一OD均满足时间窗约束,旅行时间均处于可接受范围内,而线网费用仅略高于无时间窗情形。时间窗参数δ=2.5时,接驳线网中单个OD的最大绕行比和出行时间有较好的表现;时间窗参数δ≥4.3时,接驳线网接近无时间窗约束的情况。算例结果证明了模型和算法的有效性。图33幅,表18个,参考文献92篇更多还原

【Abstract】 With the continuous development of public transport,the cooperative operation of rail transit and rail transit and bus transport is the key to the integrated development of public transportation,which can maximize the advantages of the both,improve the coverage and service quality of urban public transportation network,and enhance the attractiveness of public transportation to passengers.Based on the FBNDP problem with many-to-many demands of passenger flow,the optimization of urban rail transit related to feeder bus network with time window constraints(FBNDP-TW)is proposed in the thesis.The main contents of this thesis are as follows:(1)To ensure the service quality of the feeder bus network,the total travel time of passenger flow OD in the feeder bus network is strictly limited.A time window function based on the OD ideal travel time is constructed through the introduction of the time window parameter threshold δ to set the differential time window constraint OD by OD.In the problem of classic feeder bus network design,a travel time window constraint is added,and the optimization model of the urban rail transit related to feeder bus network is constructed with the objective of minimizing the generalized travel costs of passengers and operation costs of public transport operators,and then feeder stations,optimal feeder bus routes and operating frequency of the optimized feeder bus network lines can be obtained.(2)Based on the framework of genetic algorithms,the optimization algorithm of urban rail transit related to feeder bus network with time window is designed.To keep the service quality of lines stable and meet the time window requirement,a diversified neighborhood crossover operation composed of three rules is designed,namely,direct connection,forward connection and adjacent connection.Distinguish the situations in which the time window constraint is satisfied or not,and choose different strategies for combining crossover rules to effectively guide the evolutionary direction of the offspring individuals,which enables points in the network meet the time window requirement as soon as possible,and efficiently solves the problem.(3)Taking the standard feeder bus network as a case study,the optimization model and algorithm for urban rail transit related to feeder bus network with time windows are verified.Results show that the time window constraint has the ability to significantly reduce the detour ratio of a single OD,each OD satisfies the time window constraint,the travel time is within the acceptable range while network costs is slightly only higher than the case without the time window constraint.When the time window parameter δ =2.5,the maximum detour ratio and travel time of OD in the feeder network have a good performance.When the time window parameter δ ≥4.3,the feeder bus network is close to the situation without the time window constraint.The results demonstrate the validity of the model and algorithm.There are 33 figures,18 tables and 92 references in this thesis.更多还原