摘要
工程设计过程可看作是一约束确定与满足过程 ,柔性装配系统的优化设计也是一约束优化问题 .遗传算法是一种广泛应用于复杂工程技术系统设计的有效方法 ,但用遗传算法求解约束优化问题比较困难 .在人机一体优化概念的基础上 ,讨论了如何通过优化开始前的约束预处理、优化过程中的约束处理以及建立变异库给变异操作一个进化方向 ,来将以约束形式表达的领域知识融合到优化搜索过程中 ,从而提高了遗传算法的优化性能 ,并在柔性装配系统中成功地实现了人机一体优化设计 .
A design of an engineering project is a process of constraint specification and satisfaction. An optimal design for a flexible assembly system can be regarded as a constrained optimization problem. The method of genetic algorithm is efficient and is used widely in the design of the complex systems in engineering. However, some additional difficulties exist in the process of using genetic algorithm to solve constrained optimization problems. The present paper discusses how to combine the human knowledge into the searching process through the constraints pre_handled and mutation library built with mutation operators. Thus, the optimal performances of the genetic algorithm are improved. Finally, an optimization is successfully used in the design of a flexible assembly system.
出处
《华南理工大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2001年第10期23-27,共5页
Journal of South China University of Technology(Natural Science Edition)
基金
国家自然科学基金资助项目 (5 9775 0 81)
广东省自然科学基金资助项目 (990 5 45 )~~