地铁货运系统运作的列车调度方式建模

地铁货运系统作为地下物流系统一种很有前景的技术形式,充分利用城市既有地铁网络基础设 施,建设与城市轨道客运协同运作的地铁货运具有较高的可行性。客货分离式和拖挂式是两种主要的地铁 货运模式,由于地铁客货站点分布不同,客货列车的协同运行方式实际上是一种站点确定条件下的快慢车行 车调度问题。以满足地铁货运需求且对客运影响最小为目标,分别对两种模式的地铁货运系统构建列车调 度双层规划模型,并设计双层嵌套遗传算法进行求解。以 X市环线地铁线路为案例,利用 BP神经网络预测 未来20年的区域货运需求量,并基于此计算两种模式的列车调度排班结果。结果表明,两种模式单趟运行 时间相近,客货分离式的承载力更大,但在货运需求较小时空置率较高,易造成运能浪费;拖挂式的货运平均 等待时间较少,货物疏散效率更高,对地铁客运影响更小,但排班灵活性较差。

OnModelingTrainSchedulingModeofMetroFreightTransportationSystem

Metro freight transportation (MFT) system is a promising form of underground logistics system. It is feasible to make full use of the existing urban metro network infrastructure and build the metro freight system which is operated in coordination with the urban rail passenger transport system. Passenger-cargo separated type and trailer type are the two main modes of MFT. Due to the different distribution of passenger and cargo stations, the coordinated operation mode of passenger and cargo trains is actually a scheduling problem of fast and slow trains with determined stations. A two-layer planning model for train scheduling was constructed for the two modes of MFT with the objective of satisfying the demands for metro freight and minimizing the impact on passenger transport, and a nested genetic algorithm was designed to solve the problem. Then, the train scheduling results of the two modes were calculated based on the regional freight demand data for the next twenty years calculated by BP neural network under the background of the metro loop line in X city. The results show that (1) the two modes have similar one way operational time; (2) The passenger-cargo separated type has larger carrying capacity, while the vacancy rate is high when the freight demand is small, which is easy to cause waste of transport capacity; (3) The average waiting time of the trailer type is less, which evacuation efficiency is higher; (4) The passenger-cargo separated type has a greater impact on the capacity of the subway lines, while the scheduling flexibility of the trailer type is relatively poor.