模拟退火粒子群算法在动态单元布局中的应用  

Application of Particle Swarm Algorithm Based on Simulated Annealing in Variable Cellular Facility Layout Problems

作　　者：祝恒云[1,2] 叶文华[1] 

机构地区：[1]南京航空航天大学江苏省精密与微细制造技术重点实验室,南京210016 [2]南京工业职业技术学院,南京210046

出　　处：《中国机械工程》2009年第2期181-185,共5页China Mechanical Engineering

摘　　要：通过对单元化制造系统规划中设备布局特点以及单元零件工艺流程的分析,建立了基于动态单元的车间设备布局模型。为解决设备布局这一典型的非多项式(non-polynomial,NP)难题,提出了一种新的模拟退火粒子群算法,该算法将粒子群算法与模拟退火方法有机结合,在保留粒子群算法具有收敛速度快等优点的同时,增强了算法的全局搜索能力。通过实例仿真,进一步验证了该方法的有效性和实用性。Through analyzing some key features of the facility layout in cellular manufacturing design and the components＇ process procedure, a model of facility layout under variable demand over the product life cycle was built. To solve the facility layout problem, a new particle swarm algorithm based on simulated annealing （SAPSO） was formulated. The SAPSO can keep the fast convergence speed of standard particle swam algorithm and can reach the global optimum solutions. The computational results show that the SAPSO can obtain satisfactory solutions within reasonable time.

关 键 词：设备布局 变需求 模拟退火粒子群算法 单元化制造 

分 类 号：TP166[自动化与计算机技术—控制理论与控制工程]