基于改进NSGA-Ⅲ的内河集装箱船舶配载多目标优化  

Improved NSGA-Ⅲ in multi-objective optimization of stowage planning for inland container ship

作　　者：赵雅洁 李俊[1] 肖笛 温想 ZHAO Yajie;LI Jun;XIAO Di;Wen Xiang(School of Automobile and Traffic Engineering,Wuhan University of Science and Technology,Wuhan 430081,China)

机构地区：[1]武汉科技大学汽车与交通工程学院,湖北武汉430070

出　　处：《中国航海》2023年第3期153-162,共10页Navigation of China

基　　金：湖北省教育厅科学技术研究计划项目(Q20211110)。

摘　　要：内河集装箱运输差异化特征导致船方配载决策时考虑多目标优化,为满足船舶运输经济性和适航性需求,以优化船舶堆栈占用数量、阻塞箱数量、稳性高度、横倾角值及纵倾值为目标,构建内河集装箱船舶配载多目标优化模型。为实现多目标优化问题有效求解,采用灰熵并行分析法改进第三代非支配遗传算法(Non-dominated Sorting Genetic Algorithm-Ⅲ,NSGA-Ⅲ),将灰熵并行关联度作为适应度值引导算法进行精英选择。结果表明:改进后算法在求解性能表现上优于采用一般选择策略的算法,对算例参数设置具有较好鲁棒性,可为船方实际制定内河集装箱船舶配载计划提供一定决策支持。The diversity of inland container transportation requires multi-objective optimization in ship stowage planning.Aiming to satisfy the requirements of ship transportation economy and seaworthiness,a multi-objective optimization model for stowage of inland container ships is built.The model,based on NSGA-Ⅲ(Non-dominated Sorting Genetic Algorithm-Ⅲ),is developed to comprehensively optimize the number of ship stacks,number of blocking boxes,stability height,roll angle and pitch value.For effectively solving the problem,the grey-entropy parallel analysis method is used to improve the NSGA-Ⅲ algorithm,and the grey entropy parallel correlation degree is used as the guidance algorithm of fitness value for elite selection.The advantage of the algorithm and its robustness are demonstrated.

关 键 词：水路运输 船舶配载 多目标优化 第三代非支配遗传算法 灰熵并行分析 

分 类 号：U693[交通运输工程—港口、海岸及近海工程]