城市轨道交通列车运行图鲁棒性优化模型

摘要

High density traffic is interfered by transportation,which leads to the delay of urban rail transit.It could be solved by adjusting the buffer time of the train running process in order to achieve the timetable robustness during the planning level.With the constraint of trains' capacity considered and historical passenger data investigated,the actual dwell time was decided by the interactional relationship between the waiting passengers and trains.Based on the recursive relations between disturbance times of timetable,the robust timetabling optimization model was established.The objective of the model was to minimize the deviation time of timetabling.This model was solved by the optimized genetic algorithm since it belonged to nonlinear mixed integral programming model.Moreover,the model was extended by adopting the time control point strategy in order to satisfy the actual transport demand for the punctual departure time of important stations,such as large hub stations.Finally,an empirical case of Fangshan Line of Beijing subway was analyzed to certify the effectiveness of this model and the optimized genetic algorithm.%针对高密度行车因受运输干扰而导致城市轨道交通列车晚点的问题,按照在计划层的列车运行时段内调整列车缓冲时间以优化列车运行图鲁棒性的思路,考虑列车载客能力约束,基于历史客流数据和通过候车乘客与列车的交互关系确定列车的实际停站时间,然后基于列车运行图扰动时间的递推关系,建立以列车运行图扰动时间之和最小为目标的列车运行图鲁棒性优化模型;运用改进的遗传算法对属于非线性混合整数规划模型的该优化模型进行求解;另外还对只要求大型枢纽站等重要车站准点发车的实际运输需求,用时间控制点法对该优化模型进行扩展.以北京市城市轨道交通房山线为例验证了该优化模型和改进遗传算法的有效性.