基于改进Rotor模型的客货共线铁路通过能力计算方法  

Calculation Method for Carrying Capacity of Mixed Passenger and Freight Railway Based on Improved Rotor Model

作　　者：李和壁 田长海 张守帅[3] 江雨星 LI Hebi;TIAN Changhai;ZHANG Shoushuai;JIANG Yuxing(Postgraduate Department,China Academy of Railway Sciences,Beijing 100081,China;Railway Science&Technology Research&Development Center,China Academy of Railway Sciences Corporation Limited,Beijing 100081,China;School of Transportation and Logistics,Southwest Jiaotong University,Chengdu Sichuan 610031,China;School of Traffic and Transportation,Lanzhou Jiaotong University,Lanzhou Gansu 730000,China)

机构地区：[1]中国铁道科学研究院研究生部,北京100081 [2]中国铁道科学研究院集团有限公司铁道科学技术研究发展中心,北京100081 [3]西南交通大学交通运输与物流学院,四川成都610031 [4]兰州交通大学交通运输学院,甘肃兰州730000

出　　处：《中国铁道科学》2021年第3期144-155,共12页China Railway Science

基　　金：国家重点研发计划项目(2018YFB1201403);中国铁路总公司科技研究开发计划课题(J2018X002);中国铁道科学研究院集团有限公司科研开发基金资助项目(2018YJ15,2018YJ153)。

摘　　要：针对连发追踪旅客列车开行数量增加导致扣除系数法在通过能力计算上的局限性,提出1种用于计算客货共线铁路通过能力的全时迭代方法。对原始时刻表数据执行既定分层策略,确定运行图目标区段不同类型列车及天窗的铺画优先级,将通过能力最优解问题转化为平行运行图上逐层铺画开行列车并使每次铺画对目标值影响最小的最值求解问题;基于密码锁转子解锁原理提出改进Rotor模型,通过逐层非连续多元方程寻优实现无损耗迭代,求得目标区段通过能力的最优解;依托京沪铁路徐州北下发场—蚌埠区段的时刻表数据进行实例验证。结果表明:本文算法在计算通过能力时可充分考虑目标区段内各类列车条件;基于同样的数据集,纵向与枚举法、简单Rotor模型法对比,本文算法收敛耗时最短、计算效率最高;横向与扣除系数法、图解法以及路局反推法对比,本文算法平图利用率充分、性能最优;与路局实际运行图数据对比,本文算法得到的区段通过能力平均提升9列,区段空闲时间占用率平均降低11.05%。Aiming at the limitation of the deduction coefficient method in the calculation of the carrying capacity caused by the increase in the number of continuous tracking passenger trains,a full-time iterative method was proposed to calculate the passing capacity of mixed passenger and freight railway.The established hierarchical strategy was implemented for the original timetable data to determine the layout priority of different types of trains and maintenance windows in the target section of the train working diagram.The optimal solution of the carrying capacity was transformed into the maximum solution problem through the layer-by-layer layout of the running trains on the parallel train working diagram and minimizing the influence of each layout on the target value.Based on the principle of unlocking the cipher lock rotor,an improved Rotor model was proposed.The lossless iteration was achieved through layer-by-layer non-continuous multivariate equation optimization,and the optimal solution of the carrying capacity of the target section was obtained.An example was verified based on the timetable data of Beijing-Shanghai railway from Xuzhou North Departure Yard to Bengbu section.Results show that the proposed algorithm can fully consider all kinds of train conditions in the target section when calculating the carrying capacity.Based on the same data set,compared with the enumeration method and the simple Rotor model method,the proposed algorithm has the shortest convergence time and the highest computational efficiency.Compared with the coefficient removal method,the graphic method and the inverse method of railway bureau,the proposed algorithm has sufficient utilization rate and optimal performance.Compared with the actual train working diagram data of railway bureau,the section carrying capacity calculated by the proposed algorithm is improved by an average of 9 trains.The idle time occupancy rate of the section is decreased by 11.05%on average.

关 键 词：铁路运输 客货共线铁路 通过能力计算方法 列车运行图 扣除系数法 Rotor模型 

分 类 号：U292.5[交通运输工程—交通运输规划与管理]