基于改进免疫遗传算法的汽车零件排样  

Automobile Parts Packing Based on Improved Immunogenetics Algorithm

作　　者：闫嘉[1,2] 李林峰 林毓培 段书凯 YAN Jia;LI Linfeng;LIN Yupei;DUAN Shukai(College of Artificial Intelligence,Southwest University,Chongqing 400715,China;National&Local Joint Engineering Research Center of Intelligent Transmission and Control Technology(Chongqing),Chongqing 400715,China)

机构地区：[1]西南大学人工智能学院,重庆400715 [2]智能传动和控制国家地方联合工程研究中心(重庆),重庆400715

出　　处：《西南大学学报（自然科学版）》2023年第5期204-214,共11页Journal of Southwest University(Natural Science Edition)

基　　金：国家重点研发计划项目(2018YFB1306600).

摘　　要：为提高汽车零件排样的板材利用率,对汽车零件的排样技术进行研究.汽车零件大多是不规则零件,因此其排样属于二维不规则排样问题范畴.在智能优化算法求解二维不规则排样领域,遗传算法容易产生早熟现象,并且局部寻优能力较差,而免疫算法具有很强的局部寻优能力.结合遗传算法与免疫算法的全局与局部寻优能力,提出一种改进的免疫遗传算法.该算法的遗传选择操作加入个体浓度计算,操作步骤和变异概率可以自适应调整,从而有效避免早熟现象.自适应计算免疫选择个体数以及免疫克隆个体数以提高算法的收敛性能.分别使用ESICUP网站提供的公开数据和实际汽车零件专用数据对提出的免疫遗传算法的性能进行测试,实验结果表明,和现有的几种算法相比,本研究提出的免疫遗传算法能获得更高的材料利用率.To improve the plate utilization rate of automobile parts packing,the packing technology of automobile parts is studied.Most of the automobile parts are irregular,so this problem belongs to the category of two-dimensional irregular packing problem.In the field of intelligent optimization algorithms to solve the two-dimensional irregular packing,the genetic algorithm is liable to produce premature phenomenon and poor in local optimization ability,while immune algorithm has strong local optimization ability.In this paper,an improved immunogenetic algorithm is proposed by leveraging of combining the global and local optimization capabilities of the genetic algorithm and immune algorithm.The genetic selection operation of the algorithm adds the individual concentration calculation,and the operation steps and mutation probability can be adaptively adjusted to avoid the premature phenomenon effectively.The number of immune-selected individuals and immune-cloned individuals are calculated adaptively to improve the convergence performance of the algorithm.The performance of the proposed immunogenetic algorithm was tested using the public data provided by ESICUP website and the actual automobile parts private data,respectively.The experimental results show that the proposed immunogenetic algorithm can achieve higher material utilization rate compared with several existing algorithms.

关 键 词：二维不规则排样 临界多边形 汽车零件 智能优化算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]