• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2013, Vol. 35 ›› Issue (1): 137-141.

• 论文 • 上一篇    下一篇

一种改进的协同优化算法及其应用

李冬琴1,2,杨永祥1,陈智同2   

  1. (1.江苏科技大学船舶与海洋工程学院,江苏 镇江 212003;
    2.江苏现代造船技术有限公司,江苏 镇江 212003)
  • 收稿日期:2011-08-15 修回日期:2011-11-20 出版日期:2013-01-25 发布日期:2013-01-25
  • 作者简介:李冬琴(1979),女,湖北宜昌人,博士,副教授,研究方向为新船型设计和船舶多学科设计优化研究。
  • 基金资助:

    江苏省基础研究计划(自然科学基金)资助项目(BK2012696);江苏省自然科学基金资助项目(BK2009722);江苏高校优势学科建设工程资助项目(PAPD);2012年度江苏政府留学奖学金资助

An improved collaborative optimization algorithm and its application

LI Dongqin1,2,YANG Yongxiang1,CHEN Zhitong2   

  1. (1.School of Naval Architecture & Ocean Engineering,
    Jiangsu University of Science and Technology,Zhenjiang 212003;
    2.Jiangsu Modern Shipbuilding Technology,LTD.,Zhenjiang 212003,China )
  • Received:2011-08-15 Revised:2011-11-20 Online:2013-01-25 Published:2013-01-25

摘要:

针对协同优化方法收敛困难、优化效率低的问题,提出了一种改进的协同优化算法—ICO算法。通过引入自适应松弛因子将一致性等式约束转化为不等式约束,同时建立混合惩罚函数,将系统级约束优化问题转化为无约束优化问题,ICO算法较好地克服了传统协同优化算法难于收敛的缺点。标准算例实验结果表明,ICO算法能够有效提高优化的稳定性、可靠性和计算效率。优化结果显示了协同优化算法解决海洋供应船的设计优化问题的有效性,为解决更为复杂工程系统的设计优化问题奠定了基础。

关键词: 多学科设计优化, 协同优化, 海洋供应船

Abstract:

Aiming at the convergence difficulties and low optimization efficiency of Collaborative Optimization (CO), an improved collaborative optimization algorithm (ICO) is proposed in this paper. By introducing the adaptive relaxation factor to let the restriction of consistency equality be transformed to the restriction of nonequality as well as by constructing the mixed penalty function to convert the systemlevel constraint optimization problem into unconstraint optimization problem, the improved algorithm can overcome the disadvantages of convergence difficulties of traditional collaborative optimization. Experimental results show that the stability, reliability and computing efficiency of ICO are better than CO, and lay a foundation for solving more complex engineering system.

Key words: multidisciplinary design optimization;collaborative optimization;offshore supply vessel