铁路编组站阶段计划重调度优化方法

Optimization Method for Rescheduling Stage Plan of Railway Marshalling Stations

铁路编组站阶段计划是车站作业调度和资源分配的核心计划,但到达列车晚点、作业进度异常、调机突然故障等严重扰动事件容易导致既定计划无法执行,需要对失效计划重调度。基于最小摄动动态累积调度理论,以重调度前后的阶段计划之间延续性最好为主要目标、作业效率最高为次要目标,建立编组站阶段计划重调度的字典序多目标优化模型。模型中考虑作业执行间的逻辑、作业执行间的时态、资源容量、车流接续、配流满轴等约束,并通过建立资源停用、到达车辆波动、作业状态变化等扰动事件下模型动态调整的产生式规则,实现模型针对不同扰动事件的响应。采用约束传播和多点结构性搜索的混合算法求解各目标函数对应的分层模型,并通过实例验证重调度方法的可行性和算法的高效性。

Stage plans of railway marshaling stations are the core of station operation scheduling and resource allocation.However,some serious disturbance events can easily lead to the inability to implement some established plans,such as train arrival delay,abnormal operation progress,and sudden failure during machine adjustment.It is necessary to reschedule these failure plans.Based on the minimum perturbation dynamic cumulative scheduling theory,a lexicographic multi-objective model for rescheduling stage plans was built,with the best continuance between stage plans before rescheduling and after rescheduling as the main goal and the highest operation efficiency of marshaling stations as the secondary goal.The model considered the logic between job executions,the temporal state during job executions,resource capacity,train flow continuation,and full axis of flow allocation.The production rules of dynamic model adjustment under various disturbance events such as resource deactivation,arrival train fluctuation,and operation status changes were established to implement the model’s response to different disturbance events.Then a mixed algorithm combining constraint propagation and multi-point search was adopted to iteratively solve each hierarchical model corresponding to each objective function.The feasibility of the rescheduling method and the algorithm efficiency were verified by examples.