基于遗传算法的多规格管材或型材的优化下料  

Optimum cutting of multi-size stock tubing or profile based on genetic algorithm

作　　者：刘在良 翁旭辉 王静 夏小浩 Liu Zailiang;Weng Xuhui;Wang Jing;Xia Xiaohao(Zhejiang International Maritime College, Zhoushan, Zhejiang 316021, China;Yangfan Group CO.TD)

机构地区：[1]浙江国际海运职业技术学院,浙江舟山316021 [2]扬帆集团股份有限公司

出　　处：《计算机时代》2018年第12期67-70,74,共5页Computer Era

摘　　要：在船舶建造中存在大量的管材或型材需求,这些材料的一维优化下料(一维套料)问题一直在被研究。文章提出了一种基于遗传算法的求解方法,建立了数学模型,给出了编码解码方案和交叉变异方法等,并结合实际情况提出一种方法利用近似优化算法来修复无效基因,同时还在代代相传中采用精英保留策略尽可能保留每代最优解,有利于加快收敛。实例结果表明该算法的有效性,符合预期目标。There are a large number of pipe or profile requirements in ship construction,one-dimensional optimum cutting(one-dimensional nesting)of multi-size stock materials has been being studied.This paper presents a solution method based on genetic algorithm,establishes a mathematical model,gives the coding and decoding scheme,and cross and mutation methods,according to the actual situation,proposes a method to repair invalid genes by using approximate optimization algorithm.At the same time,elite retention strategies are used to preserve the optimal solution of each generation as far as possible in the process of generation-to-generation transmission,which is helpful to speed up the convergence.The example shows that the algorithm is effective and meets the expected goal.

关 键 词：一维优化下料 遗传算法 优化算法 一维套料 

分 类 号：U671.2[交通运输工程—船舶及航道工程]