摘要
为提高城市轨道交通的运营效率,使运量-运能之间能具有更好的匹配关系,研究列车的实时调度问题。在分析旅客需求特征和行车条件的基础上,以列车的运行时间、停站时间、发车时刻为决策变量,以旅客出行时间最小化为目标构建混合整数非线性规划模型,提出序列二次规划和迭代凸规划两种算法进行求解。最后,以广州地铁8号线为例进行分析,算例表明该模型在列车实时调度方面具有较好的实用性,基于迭代的凸规划算法可显著提高大规模问题的求解速度。
In order to improve the operational efficiency and better match between the passenger demand and the traffic capacity, the real-time train scheduling problem for an urban rail transit line is discussed. Based on the analysis of passenger demand characteristics and train operation conditions, a mixed integer nonlinear programming model which regards running time, stop time and departure time as the decision variable is established to minimize the total travel time of passengers. The sequential quadratic programming and the iterative convex programming approaches are introduced to find the best timetable. Finally, a numerical example taking the relevant information of Guangzhou metro line 8 is given to demonstrate the effectiveness of the model proposed and the better convergence speed of the latter algorithm.
出处
《交通科技与经济》
2017年第4期30-35,72,共7页
Technology & Economy in Areas of Communications
基金
五邑大学教学研究和改革项目(JX2016014)
广东省高等教育教学研究和改革项目(32041002)
关键词
城市轨道交通
实时调度
混合整数非线性规划模型
序列二次规划
迭代凸规划
urban rail transit
real-time train scheduling
mixed integer nonlinear programming model
sequential quadratic programming（SQP）
iterative convex programming（ICP）
