协同优化算法的改进  

IMPROVEMENT OF COLLABORATIVE OPTIMIZATION

作　　者：韩明红[1] 邓家禔[1] 

机构地区：[1]北京航空航天大学机械工程及自动化学院,北京100083

出　　处：《机械工程学报》2006年第11期34-38,共5页Journal of Mechanical Engineering

基　　金：国家自然科学基金(50275005);国防基础科研(K1203010804)资助项目

摘　　要：针对协同优化算法是多学科设计优化方法中应用最广、效果最好的MDO算法,但是在应用中存在计算困难的现状,分析了引起协同优化算法计算困难的原因,提出一种改良的协同优化算法——ICO多学科设计优化方法。改进学科一致性约束,增加系统级罚函数,采用快速启动方法,较好地克服了协同优化算法存在计算困难的缺点。标准算例试验结果表明,ICO多学科设计方法能够有效提高算法的稳定性、可靠性和计算效率。Collaborative optimization （CO） is the most widely used MDO method, which has good effect in application, however some serious computational difficulties are found in its application. Reasons that cause computational difficulties in collaborative optimization are analyzed and a new improved collaborative optimization method （ICO） is presented. Improved subsystem consistency constraint, system-level penalty function, and a quick-start strategy are used in ICO. These measures overcome the computational difficulties of CO better. Experimental results show that the robustness, reliability and computing efficiency of ICO are higher than CO.

关 键 词：优化 计算机辅助设计 改进 协同优化算法 多学科设计优化 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]