基于虚拟编组的城轨列车时刻表优化方法  

A Timetable Optimization Method for Urban Train Transit Based on Virtual Coupling

作　　者：李岩 巩亮[1] 许得杰[1] 潘星 胡晨皓 LI Yan;GONG Liang;XU Dejie;PAN Xing;HU Chenhao(School of Traffic and Transportation,Lanzhou Jiaotong University,Lanzhou 730070,China)

机构地区：[1]兰州交通大学交通运输学院,兰州730070

出　　处：《交通信息与安全》2024年第3期74-84,共11页Journal of Transport Information and Safety

基　　金：国家自然科学基金项目(72261025);甘肃省高等学校创新基金项目(2021B-103);甘肃省教育厅双一流重大科研项目(GSSYLXM-04)资助。

摘　　要：针对高峰时段城市轨道交通客流与运力不匹配问题,考虑城市轨道交通客流时空分布特征、交通过饱和状态和车场车底数量限制,提出了基于虚拟编组的城轨列车时刻表优化方法。构造动态客流累计需求函数,预测不同时段客流量;考虑乘客需求、发车间隔、运行时间、车底数量、车底接续等约束条件,以列车在首站的发车时刻和各车次的编组方案为决策变量,以乘客平均等待时间与列车走行里程最小化为优化目标,建立了基于虚拟编组的城轨列车时刻表优化模型。针对原问题包含大量耦合约束条件,利用拉格朗日松弛算法将耦合性约束吸收至目标函数,将原问题分解为2个独立路径的子问题,降低问题的复杂度;再利用商业求解器求解子问题的下界解,并设计启发式算法求解子问题的上界可行解,得到原问题解的上下界。以上海地铁某线路为算例进行验证,结果表明:在高峰时段,所提动态客流累计需求函数与客流实际到达规律拟合度较高;固定编组模式下,非均匀发车时刻表相比于均匀发车时刻表,可降低24.15%的乘客平均等待时间和51.73%的滞留乘客等待时间;而所提虚拟编组列车时刻表相比于固定编组模式下非均匀发车时刻表,不仅可减少0.33%的列车运行里程,还可进一步减少16.95%的乘客平均等待时间和6.03%的滞留乘客等待时间。To solve the mismatch between train capacity and demand during peak hours,a timetable optimization method for urban train transit based on virtual coupling technical is proposed,incorporating spatiotemporal characteristics of passenger flow,oversaturation of trains during peak hours,and the limitation of the number of rolling stocks.A dynamic passenger flow cumulative demand(PFCD)function is proposed to pedict the passenger flow at different hours.Then,the schedule optimization model for urban rail transit based on the virtual coupling is established,in which,the departure time of trains at the first station and the marshaling scheme of each train are decision variables and the average waiting time(AWT)of passengers and the train travel mileage(TTM)are minimized under constraints such as passenger demand in different hours,departure interval,running time,number of rolling stocks,rolling stock circulation,etc.Lagrangian relaxation is introduced to reduce the complexity of the problem by absorbing the coupling constraints into the objective,and the original problem is decomposed into two independent subproblems.By using a commercial solver and the designed heuristic algorithm,the lower bound and upper bound of the problems are found.A metro line in Shanghai Metro is employed for demonstration,and the results show that:①the proposed dynamic PFCD function fits the arrival pattern of passengers well during the peak hours;②compared with the uniform departure schedule,the non-uniform departure(non-UD)schedule under the fixed train composition(FTC)mode can reduce the AWT of passengers by 24.15%and the waiting time of stranded passengers by 51.73%;③compared with the non-UD schedule under the FTC mode,the train timetable based on virtual coupling can reduce not only the train running kilometers by 0.33%but also the AWT of passengers and the waiting time of stranded passengers by 16.95%and 6.03%,respectively.

关 键 词：城市交通 城市轨道交通 列车时刻表 拉格朗日松弛算法 动态客流需求 虚拟编组 

分 类 号：U292.4[交通运输工程—交通运输规划与管理]