摘要
后优化分析技术定量地研究当优化参数发生变化时,相应的优化结果会发生怎样的改变。过去的研究多集中在对优化结果的敏感度分析(一阶近似),所得到的结果只在当前参数的微小扰动范围内有效。用Kriging模型进行优化结果的插值近似(相对于优化参数),从而可以提供整个参数空间中优化结果的全貌。用简单的算例与一阶近似、二次响应面近似进行了比较,同时为了解决临界约束集发生改变时,目标函数突变对Kriging模型近似精度带来的影响,提出一种分片插值方案,较大地提高了Kriging的近似精度。
The post-optimality analysis techniques quantitatively approach how the optimum solutions change with respect to optimization parameters without re-optimizing the problem.In recent years,these techniques have been used by several multidisciplinary design optimization(MDO)frameworks.This paper is focused on creating global surrogate models in the whole parameter space,especially by the Kriging method.Firstly,Kriging model is directly used to approximate the post-optimality relations of a simple mathematical optimization problem and are compared with the widely used second order polynomial response surface method.Secondly,a piecewise Kriging scheme is proposed in order to reduce the influence of possibly saltations of the optimum sloutions on the accuracy of Kriging model as a result of changes of the active constrains set.Finally,a simple structural optimization problem is used to demonstrate the improvement on the accuracy.
出处
《机械设计与研究》
CSCD
2004年第5期10-13,共4页
Machine Design And Research
基金
国家自然科学基金资助项目(50075065)
关键词
后优化分析
KRIGING模型
多学科优化
Post-optimality analysis
Kriging model
Multidisciplinary Design Optimization(MDO)
