考虑列车解编时间随机性与出发时刻柔性约束的相邻技术站间协同配流研究

Collaborative Railcar Flow Allocation between Adjacent Technical Stations Considering Randomness of Train Disintegration and Marshalling Operation Time and Flexible Constraint of Departure Time

为提高技术站货车周转效率,增强车流接续计划的可靠性,研究考虑列车解编时间随机性与出发时刻柔性约束的相邻技术站间协同配流问题,在尽可能减少列车出发时刻调整量条件下,以所有车辆在两个技术站的总停留时间最小为目标,构建相邻技术站间协同配流机会约束规划模型,并设计基于到发车流接续差的邻域搜索算法进行求解。该算法从列车解编顺序调整的本质出发,计算理想接续车流与实际接续车流之间的车流接续差,据此得到解编调整点,结合列车解编顺序调整所对应的时间变化量构造邻域,优化解编顺序。算例结果表明:考虑列车出发时刻柔性以及站间协同优化,所有车辆在站总停留时间减少了86.96 h,阶段内发出列车数增加2。设计的算法可以在短时间内收敛至最优解,得到可靠性较高的配流方案。

In order to improve the turnover efficiency of freight cars at technical stations and enhance the reliability of the car flow joining plan,the collaborative railcar flow allocation of freight trains between adjacent technical stations was studied considering the randomness of train disintegration operation time and marshalling operation time and flexible constraint of departure time.Under the condition of minimizing the adjustment amount of the train departure time,the chance constraint programming model of collaborative railcar flow allocation between adjacent technical stations was constructed with the goal of minimizing the total residence time of all cars at the two technical stations,while the neighborhood search algorithm based on the car flow joining difference between the arrival train and the departure train was designed to solve the problem.Starting from the essence of the adjustment of train disintegration sequence and marshalling sequence,this algorithm calculated the car flow joining difference between ideal car flow joining and actual car flow joining,to obtain the train disintegration sequence and marshalling sequence adjustment points.The neighborhood was constructed by combining the time change caused by the adjustment of train disintegration sequence and marshalling sequence,to optimize the disassembly sequence and the marshalling sequence.The example results show that considering the flexibility of the train departure time and the collaborative optimization between stations,the total residence time of all cars at the stations is reduced by 86.96 hours,with the number of successful departure trains increased by 2.The designed algorithm can converge to the optimal solution in a short time and obtain a more reliable railcar flow allocation scheme.