基于双离散变量遗传算法的洗舱站选址研究  

Site selection for tank cleaning station based on genetic algorithm with multiple discrete variables

作　　者：王新月 马晓凤[3,4] 文元桥[3,4] WANG Xinyue;MA Xiaofeng;WEN Yuanqiao(Laboratory of Transport Pollution Control and Monitoring Technology,Beijing 100028,China;China Automotive Technology Research Center Co.,Ltd.,Tianjin 300300,China;Intelligent Transportation Systems Research Center,Wuhan University of Technology,Wuhan 430063,China;National Engineering Research Center for Water Transport Safety,Wuhan University of Technology,Wuhan 430063,China)

机构地区：[1]交通排放控制监测技术实验室,北京100028 [2]中国汽车技术研究中心有限公司,天津300300 [3]武汉理工大学智能交通系统研究中心,湖北武汉430063 [4]武汉理工大学国家水运安全工程技术研究中心,湖北武汉430063

出　　处：《中国航海》2024年第1期121-130,共10页Navigation of China

基　　金：国家自然科学基金(51678461);交通排放控制监测技术实验室开放课题((2024)JH-F005)。

摘　　要：针对洗舱站选址与规模问题,首先引入“港口洗舱热度值”的概念,确定各个港口潜在的洗舱市场;其次以系统总成本最低为目标函数,以洗舱站位置与规模为决策变量,建立洗舱站选址模型,采用双染色体遗传算法求解模型,利用基于港口洗舱热度值的初始解选择策略、自适应交叉变异概率、精英保留策略对算法进行改进;最后以长江部分航段为案例进行实例分析。结果表明:求解大规模问题时,遗传算法在求解效率上优于Lingo求解器;选址结果集中在洗舱热度值靠前的港口;所选案例建设6座洗舱站时空载距离最小,为250千米/(艘·年)左右。The concept of"cabin washing heat value(CWHV)"for ports is introduced to reflect the eagerness of a port for cabin washing,and the cabin washing heat values of interested ports are determined.The location selection model for cabin washing station is built with the objective function for lowest overall cost.The location and the scale of the cabin washing station are taken as the decision variables.A modified dual chromosome genetic algorithm is used to solve the model.The modification includes CWHV-based initial selection strategy,adaptive crossover probability and elite retention strategy.The planning of cabin washing stations for a group of ports on the Yangtse River is carried out as an example.The solution was 6 cabin washing stations distributed at ports with higher CWHV,which guarantees minimum deadhead kilometers,about 250 km/(ship:year).The process demonstrated the efficiency advantage of genetic algorithm over Lingo solver for large scale problems.

关 键 词：洗舱站 选址 离散变量 遗传算法 

分 类 号：U651.1[交通运输工程—港口、海岸及近海工程]