| 基于动态罚函数的铁路车流分配与径路优化模型 |
| |
| 引用本文： | 薛锋,刘泳博,户佐安,陈逸飞.基于动态罚函数的铁路车流分配与径路优化模型[J].西南交通大学学报,2022,57(5):941-948, 959. |
| |
| 作者姓名： | 薛锋 刘泳博 户佐安 陈逸飞 |
| |
| 作者单位： | 1.西南交通大学交通运输与物流学院，四川 成都 6117562.西南交通大学唐山研究生院，河北 唐山 0630003.西南交通大学综合交通大数据应用技术国家工程实验室， 四川 成都 611756 |
| |
| 基金项目： | 国家自然科学基金（61203175）；四川省自然科学基金（2022NSFSC0471）；四川省科技计划（2021YJ0067，2021YJ0077） |
| |
| 摘 要： | 为解决铁路车流分配与径路优化模型中的难约束问题，避免群智能算法在应对该问题时难以求解的不足，提出了一种基于惩罚函数的约束优化方法. 首先，在车流分配及径路优化基本模型的基础上设置虚拟弧，在目标函数中增加惩罚项的方式松弛掉模型中的弧段能力约束，同时对惩罚项中的惩罚力度和惩罚因子设计动态更新的策略；然后，将改进灰狼算法（improved grey wolf algorithm，IGWO）应用于车流分配与径路优化模型的求解；最后，结合某一地区的路网数据，对改进前、后的模型和算法进行对比分析. 算例结果表明:与改进前的模型相比，引入惩罚项之后，IGWO可以在限定的范围内找到满足弧段能力约束的可行解；与灰狼算法（gray wolf algorithm，GWO）相比，IGWO计算所得的配流方案使OD （origin-destination）货流的平均绕行率和货物总走行公里数分别下降了2.6%和5.2%.
|
| 关 键 词： | 铁路运输 车流分配 径路优化 惩罚函数 约束优化 改进灰狼算法 |
| 收稿时间： | 2021-03-29 |
Railcar Traffic Distribution and Route Optimization Model Based on Dynamic Penalty Function |
| |
| XUE Feng,LIU Yongbo,HU Zuoan,CHEN Yifei.Railcar Traffic Distribution and Route Optimization Model Based on Dynamic Penalty Function[J].Journal of Southwest Jiaotong University,2022,57(5):941-948, 959. |
| |
| Authors: | XUE Feng LIU Yongbo HU Zuoan CHEN Yifei |
| |
| Affiliation: | 1.School of Transportation and Logistics, Southwest Jiaotong University, Chengdu 611756, China2.Graduate School of Tangshan, Southwest Jiaotong University, Tangshan 063000, China3.National Engineering Laboratory of Integrated Transportation Big Data Application Technology, Southwest Jiaotong University, Chengdu 611756, China |
| |
| Abstract: | In order to solve the constraint difficulty in the railway traffic distribution and route optimization model, and the inadequacy of swarm intelligence algorithm, a constraint optimization method based on penalty function is proposed. First, a virtual arc is set on the basis of the basic model of traffic distribution and route optimization, and the arc segment capability constraints in the model are relaxed by adding a penalty term to the objective function. Meanwhile, a dynamic update strategy is designed for the penalty intensity and penalty factor of the penalty term. Then, the improved grey wolf algorithm (IGWO) is applied to the solution of traffic distribution and route optimization models. Finally, combined with the road network data in a certain area, the models and algorithms before and after improvement are compared and analyzed. The results of case study show that, compared with the model before improvement, after introducing the penalty term, IGWO can find feasible solutions that satisfy the arc capacity constraints within a limited range; compared with the grey wolf algorithm (GWO), the distribution scheme calculated by IGWO reduces the average detour rate of the OD (origin-destination) cargo flow and the total cargo kilometers by 2.6% and 5.2%, respectively. |
| |
| Keywords: | |
|
| 点击此处可从《西南交通大学学报》浏览原始摘要信息 |
|
点击此处可从《西南交通大学学报》下载全文 |
|
