摘要
针对多编组模式下地铁列车故障救援方式及运行调整计划的复杂度和精细化要求,考虑各故障场景下可选救援方式的多样性,将预定救援方式的关键要素作为输入参数,构建以列车延误时间和乘客总出行时间最小化为目标的运行图调整优化模型,设计结合变邻域搜索策略的非支配遗传算法Ⅱ,在扣车和加开备车基础上,对比取消车次、提前回段、小交路折返和灵活接续等运行调整策略的效果。结果表明:相较于就近接续策略,灵活接续策略可使编组类型匹配概率提升至少20.04%;采用取消车次、提前回段和小交路折返等运行调整策略且实施对象均为小编组列车时,可在缓解延误传播的同时降低运行调整对乘客出行的影响;在故障列车为小编组且前方停车线长度满足大编组列车停放需求的情况下,采用后序大编组列车解编救援时的运行调整效果优于列车直接救援的;相比前序/对向列车直接救援、后序/前序列车联挂等多种救援方式,后序列车直接救援得到的运行调整效果更优;设计算法中引入的变邻域搜索策略可显著提高帕累托解集的整体质量。
Regarding the complexity and precision required for metro train failure rescue schemes and rescheduling under multi-formation modes,the diversity of available rescue schemes across various failure scenarios is considered.Taking key elements of the predetermined rescue scheme as input parameters,an optimization model for timetable rescheduling is constructed with the objectives of minimizing both train delay time and total passenger travel time.The Non-dominated Sorting Genetic AlgorithmⅡintegrated with variable neighborhood search strategies is designed.Furthermore,a comparative analysis is conducted on effects of rescheduling strategies including cancellations,early depot return,minor-route turnback,and flexible connection based on train holding and additional backup trains.The results indicate that compared with the nearest connection strategy,the flexible connection strategy increases the matching probability of formation types by at least 20.04%.Applying rescheduling strategies like cancellations,early depot return,and minorroute turnback specifically to short-formation trains can mitigate delay propagation with less impact of rescheduling on passenger travel.In scenarios where a short-formation train fails and the ahead parking track is sufficiently long to accommodate a long-formation train,decoupling the following long-formation train for rescue yields better rescheduling outcomes than direct rescue.Direct rescue by the following train proves better rescheduling effects than alternative modes such as direct rescue by the preceding/opposite train or rescue via following/preceding coupling.The variable neighborhood search strategies integrated into the designed algorithm significantly enhance the overall quality of the obtained Pareto solution set.
作者
王卓
陈绍宽
曹苧木
柏赟
许浩喆
WANG Zhuo;CHEN Shaokuan;CAO Ningmu;BAI Yun;XU Haozhe(MOT Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport,Beijing Jiaotong University,Beijing 100044,China;Integrated Transport Research Center of China,Beijing Jiaotong University,Beijing 100044,China;Department of Traffic and Transportation Planning,Shenyang Urban Planning&Design Institute Co.,Ltd.,Shenyang Liaoning 110000,China)
出处
《中国铁道科学》
北大核心
2026年第1期197-209,共13页
China Railway Science
基金
国家自然科学基金资助项目(U2569203)。
