混合模拟退火算法求解单元制造系统的成本问题  

Hybrid simulated annealing algorithms for the cost problem in cellular manufacturing systems

作　　者：胡双权 王居凤 HU Shuangquan;WANG Jufeng(College of Sciences,China Jiliang University,Hangzhou 310018,China)

机构地区：[1]中国计量大学理学院,浙江杭州310018

出　　处：《中国计量大学学报》2023年第1期142-150,共9页Journal of China University of Metrology

基　　金：教育部人文社会科学研究规划基金项目(No.22YJA630088)。

摘　　要：目的:解决单元制造系统中零件加工、装配、移动成本以及机器保养与管理成本最小化问题。方法:首先在具有加工与装配单元的单元制造系统,构建一种非线性整数规划模型;针对这种复杂模型,设计了混合模拟退火算法。然后通过基于优先规则的启发式算法生成的可行初始解,替代传统模拟退火算法的随机初始解,选用以翻转的方式生成新邻域解,并采用改进前向递归算法计算目标函数值。最后在相同运行时间下,将提出的混合模拟退火算法与传统模拟退火算法和遗传算法进行了比较。结果:混合模拟退火算法的平均成本比传统模拟退火算法低25.11%~38.03%,比遗传算法低9.44%~25.64%。结论:在相同运行时间下,混合模拟退火算法在成本优化方面优于模拟退火算法和遗传算法。Aims:This paper aims to minimize the cost of the moving,processing,and installing of machine parts and the maintenance of the machines in a cellular manufacturing system.Methods:Firstly,a nonlinear integer mathematical model was proposed in a cellular manufacturing system with machining and assembly cells.Secondly,a hybrid simulated annealing algorithm was designed.A feasible initial solution generated by the priority-rule-based heuristic algorithm substituted the random initial solution generated by the traditional simulation annealing algorithm.The objective function was computed by the revised forward recursion algorithm.Results:The average cost calculated by the hybrid simulated annealing algorithm was 25.11%~38.03%lower than that of the simulated annealing algorithm;and was 9.44%~25.64%lower than that of the genetic algorithm.Conclusions:The results show that the proposed algorithm is superior to the simulated annealing algorithm and the genetic algorithm.

关 键 词：单元制造系统 优先规则 启发式算法 混合模拟退火 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]