摘要
基于标准协同优化算法,针对已有改进协同优化算法的松弛因子法和罚函数法的缺陷,引入松弛因子构造混合动态罚函数改进协同优化算法,在Isight优化软件中采用了同时具备非支配排序遗传算法和自适应模拟退火算法优点的混合算法优化系统级。将改进的协同优化算法应用到船舶结构的多目标优化设计中,对船舶机舱结构的静力学和动力学特性进行优化,得到最优解并与已有的基于动态罚函数的协同优化算法结果进行比较。优化结果表明,基于混合动态罚函数改进协同优化算法的迭代次数更少,目标值更优且学科间不一致信息更小,对于实际船舶工程上的多目标多学科结构优化有一定应用价值。
Based on the standard collaborative optimization and aimed at the limitation of existing improved collaborative optimization like the relaxation factor method and the penalty function method,the collaborative optimization algorithm was improved by introducing the relaxation factor to construct the mixed dynamic penalty function.A hybrid algorithm based on the Non-dominated Sorting Genetic Algorithm and Adaptive Simulated Annealing was proposed for system-level optimization through the Isight software.The improved collaborative optimization algorithm was applied to the multi-objective optimization design of ship structures to optimize the static and dynamic characteristics of the marine engine room.The optimal solution was obtained and compared with the results of existing collaborative optimization algorithm based on the dynamic penalty function.It was shown that the improved algorithm has fewer iterations,better target value,and smaller interdisciplinary inconsistency information,which has certain value for multi-objective and multi-disciplinary structural optimization in actual ship engineering application.
作者
郭天奇
夏益美
王福花
王德禹
GUO Tianqi;XIA Yimei;WANG Fuhua;WANG Deyu(State Key Laboratory of Ocean Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;Collaborative Innovation Center for Advanced Ship and Deep-Sea Exploration,Shanghai 200240,China;Marine Design and Research Institute of China,Shanghai 200011,China)
出处
《振动与冲击》
EI
CSCD
北大核心
2019年第20期70-76,共7页
Journal of Vibration and Shock
基金
财政部、教育部重大专项“船舶数字化智能设计系统”基金(201335)
工信部高技术船舶科研基金([2016]548)
关键词
协同优化算法
混合动态罚函数
混合算法
船舶机舱
collaborative optimization algorithm
mixed dynamic penalty function
hybrid algorithm
marine engine room
作者简介
第一作者:郭天奇,男,硕士,1994年生;通信作者:王德禹,男,博士,教授,1963年生。
