基于近似模型的卫星动力学多目标优化  

Dynamic Multi-Objective Optimization of Satellite Based on Approximation Models

作　　者：夏昊[1] 诸成[2] 陈昌亚[3] 王德禹[1] 

机构地区：[1]上海交通大学海洋工程国家重点实验室,上海200240 [2]上海微小卫星工程中心,上海200050 [3]上海卫星工程研究所,上海200240

出　　处：《上海航天》2013年第5期48-52,共5页Aerospace Shanghai

摘　　要：对基于近似模型的卫星结构动力学多目标优化进行了研究。给出了响应面模型、Kriging模型和径向基函数(RBF)模型三种近似模型的原理。基于某卫星结构有限元模型建立了优化的三种近似模型,给出了优化流程。分析结果表明:RBF模型的精度最高,该模型结合遗传算法的优化方法不仅显著降低了计算时间成本,而且有良好的精度。The multi-objective optimization of satellite based on approximation models was studied in this paper. The principles of the three approximation models which were response surface model, Kriging model and radial basis function （RBF） model were given out. The three approximation models were established based on the structure finite element model of some satellite, and the optimization flowchart was given out. The analysis results showed that RBF model had the best precision, and the optimization method which combined RBF model and genetic algorithm would not only cut down tl^e consuming time greatly and have good precision.

关 键 词：动力响应 多目标优化 近似模型 径向基函数 

分 类 号：O24[理学—计算数学] O32[理学—数学]