论文部分内容阅读
技术站列车编组计划编制问题属于超大规模的组合优化问题,求解难度较大。以往的研究在考虑技术站改编能力限制时基本上是采用取上限的计算方法,由此得到的最优方案有可能使得列车途经各技术站的改编能力利用不均衡,影响编制方案的实用性。本文在现有技术直达列车编组计划研究成果的基础上,综合考虑车站编组能力、解体能力、调车线容车数等影响因素,以技术站车辆集结消耗、改编消耗整体最小以及技术站改编能力均衡利用为目标函数,构建协同优化的多目标0-1规划模型,提出了基于分块编码的改进型遗传算法的优化方法。算例表明,该算法能有效地求解技术站单组列车编组计划方案,并能取得快速准确的良好效果,为车流组织人员提供可行的优选方案。
Technical station train compilation planning problems belong to the large-scale combinatorial optimization problems, the solution is more difficult. In the past, the calculation method based on the upper limit was basically taken into account when considering the limitation of technical station adaptation. The optimal solution obtained from this study may make the utilization ability of adaptation ability of trains passing various technical stations unbalanced and affect the practicability of the preparation scheme. Based on the research achievements of the prior art train compilation project, this paper synthetically considers the influencing factors such as the capacity of marshalling station, disintegration capacity and the capacity of shunting trolley in order to minimize the total consumption of vehicles and to adapt the technical stations Balanced utilization as the objective function to build a collaborative optimization multi-objective 0-1 programming model, and proposes an improved genetic algorithm based on block coding. The example shows that this algorithm can effectively solve the single station train grouping plan in technical station, and can obtain fast and accurate good results and provide feasible solutions for traffic flow organizers.