旅客导向的高速铁路列车运行与动车组运用一体化调整理论与方法研究

High speed trains are inevitably deviated from the original timetable in daily operations due to uncertain factors. Since the high speed railway network is expanding and the speed and density of trains are becoming higher, train rescheduling in a disruption has been one of the key issues that need to be solved urgently in railway dispatching field. This project focuses on railway timetable and multiple units rescheduling on high speed railway network in a disrupted situation according to the passenger demand. The main innovative contents are as follows: 1) The changing feature of passenger travel volume under various types of disruptions is analyzed based on the historical data in the railway ticket system by data mining; 2) The passenger route choice behavior in a disrupted situation is formulated by the shortest path approach that combining the Dijkstra algorithm and branch and bound algorithm; 3) Based on the event-activity network, a mixed integer linear programming model is formulated to reschedule the timetable and multiple units simultaneously. The model is solved by a space-time decomposition algorithm; 4) Train rescheduling strategies that minimize the total generalized travel cost are obtained by iteratively solving the passenger route choice model and the timetable and multiple units rescheduling model. This research can help passengers to select the best travel paths in a disrupted situation, and assist the involved dispatchers to handle disrupted trains.

高速列车在日常运营过程中受不确定因素的影响偏离计划运行图，随着高速铁路网规模的扩大及行车速度、密度的增加，干扰下列车运行调整已成为铁路调度指挥方面亟需解决的关键问题之一。本项目基于旅客实际出行需求，对干扰下高速铁路局部路网列车运行与动车组运用调整问题进行研究。主要创新内容有：（1）挖掘铁路客票系统中大量历史数据，分析不同类型干扰下旅客出行量的变化特征；（2）采用Dijkstra算法和分支定界算法相结合的最短路方法，建模干扰下旅客出行路径选择行为；（3）基于事件-活动网络，构建高速铁路列车运行与动车组运用一体化调整混合整数线性规划模型，依据时空分解思想对模型进行求解；（4）结合实时客流量，通过干扰下旅客出行路径选择模型与列车运行和动车组运用计划一体化调整模型之间的迭代优化，获取旅客导向的列车运行调整方案。项目的研究可为干扰下旅客的出行路径选择提供指导，为调度员的实时调度决策提供依据。

高速铁路在日常运营过程中会受到各种干扰因素的影响，导致列车运行秩序紊乱、旅客出行不便。当前高速铁路干扰下的列车运行调整主要是靠调度员人为进行，导致他们工作压力大，且目前大部分的研究是分别调整列车运行图和动车组运用计划，将两者一体化研究的比较少，此外，列车运行调整大多从列车运行的角度出发，很少考虑旅客的实际出行需求。基于此，本项目结合我国高速铁路现场的实际客流数据，分析了高速铁路旅客出行路径选择行为，进而构建了列车运行图调整和旅客出行路径选择一体化优化模型，模型在调整列车运行图时考虑动车组接续约束，从而保证模型的解在动车组运用层面可行。为求解实际问题，设计了有效的分解算法对提出的模型进行求解，加快了模型的求解速度。采用我国高速铁路实际路网数据对项目所提出的模型和算法进行了测试，结果表明同时考虑动车组运用计划和列车运行图调整，可以优化动车组的运用，为铁路企业节省运营成本，此外，在列车运行调整模型中考虑旅客的出行路径选择行为能优化列车调整图的质量，降低干扰对旅客服务质量的影响。项目研究成果能为我国高速铁路干扰下的列车运行调整自动化提供理论依据，为干扰下旅客出行的路径选择提供引导，具有重要的理论意义和工程应用价值。