基于免疫-蚁群算法的铁路危货运输径路优化  

Transportation Path Optimization of Railway DangerousGoods Based on Immune-ant Colony Algorithm

作　　者：孔德扬 王梦杰 董雪 张利 KONG Deyang;WANG Mengjie;DONG Xue;ZHANG Li(School of Traffic and Transportation,Xi'an Traffic Engineering Institute,Xi'an 710000,China;Anshan Train Depot,China Railway Shenyang Bureau Group Co.,Ltd.,Anshan 114000,China;Meihekou Train Depot,China Railway Shenyang Bureau Group Co.,Ltd.,Meihekou 135000,China)

机构地区：[1]西安交通工程学院交通运输学院,陕西西安710000 [2]中国铁路沈阳局集团有限公司鞍山车务段,辽宁鞍山114000 [3]中国铁路沈阳局集团有限公司梅河口车务段,吉林梅河口135000

出　　处：《甘肃科学学报》2023年第5期40-49,共10页Journal of Gansu Sciences

基　　金：陕西省教育厅科研计划项目资助(21JK0748)。

摘　　要：为保证铁路危险品运输安全,优化危险品运输径路,在深入分析运输整体流程的基础上提出危险品运输特定运行径路。在考虑运输径路长度、时间、人口密度等基本因素的基础上,加入列车对数、客车扣除系数、线路坡度等适应度较高的自变量,创建以安全风险、运输时间和运输收益为3个子目标的0-1整数多目标规划模型。根据子目标函数涉及的准则层指标,使用层次分析法(AHP)确认子目标函数的权重值,将多目标模型线性转化为单目标模型。利用各技术站到达其他车站的独立车流组号,确定网络中各点相互连接情况,构建带权邻接矩阵,选择免疫-蚁群混合算法对模型进行求解,并利用沈阳和哈尔滨铁路局集团公司管内复杂跨环进行验证。结果表明:锦州站-佳木斯站危险品运输的特定径路对比最短径路安全性提升了84.59%。混合算法在20代以内快速收敛,对比免疫算法与蚁群算法径路最优值分别减少了0.09%和2.30%。In order to ensure the railway transportation safety of dangerous goods and optimize the transportation path of dangerous goods,based on the in-depth analysis of the overall transportation process,it is proposed that dangerous goods transportation can transport with a special path.On the basis of considering the length,path,time,population density,and other basic factors of transportation,the 0-1 integer multi-objective programming model with safety risk,transportation time and transportation revenue as three subgoals is established by adding the independent variables with high fitness such as train pairs,passenger train removal coefficient,and line slope.According to the criteria layer index involved in the subgoal function,using analytic hierarchy process(AHP)to confirm the weight value of the sub-objective function,and the multi-objective model is transformed into a single objective model.Using the independent traffic flow number from each technical station to other stations,determining the interconnection of each point in the network,constructing the weighted adjacency matrix,selecting the immune-ant colony hybrid algorithm to solve the model,and using the complex circularity in the Shenyang and Harbin Railway Bureau Group Company for verification.The results show that compared with the shortest path,the safety factor of the special path from Jinzhou Station to Jiamusi Station is improved by 84.59%.The hybrid algorithm converges rapidly within 20 generations.Compared with the immune algorithm and ant colony algorithm,the optimal path value is reduced by 0.09%and 2.30%.

关 键 词：交通规划 危险品运输径路 特定径路 0-1整数多目标规划模型 免疫-蚁群混合算法 

分 类 号：U292.36[交通运输工程—交通运输规划与管理]