基于Kriging自适应代理模型的气动优化方法  

Aerodynamic Optimization Method Based on Kriging Adaptive Surrogate Model

作　　者：夏露[1] 王丹[1] 

机构地区：[1]西北工业大学翼型叶栅空气动力学国防科技重点实验室,陕西西安710072

出　　处：《航空计算技术》2013年第1期13-17,共5页Aeronautical Computing Technique

基　　金：国家自然科学基金项目资助(11172242)

摘　　要：气动优化设计中,引入代理模型可以有效减少计算周期,而运用有效的插值和选样方法(自适应选样)可以大大减少建立代理模型的时间,因此提出了一种基于Kriging自适应代理模型的气动优化方法。使用Kriging方法建立代理模型,通过求解EI函数最大值得到添加样本点更新代理模型,提高了代理模型的拟合精度。针对Kriging自适应代理模型的精确性和有效性,分别进行典型函数测试分析和翼型算例验证。结果表明:基于Kriging自适应代理模型气动优化方法可以实现高效的翼型气动性能优化设计。In order to reduce the computation cycle, the Surrogate Model method is applied in the aerody- namic optimization design. Using effective interpolation and sampling methods （adaptive sampling） has been proved to be an effective reduction of the model establishing time. So, an aerodynamic optimization method based on the Kriging adaptive surrogate model is proposed in this paper. Firstly, using the Kriging method to establish surrogate model and then adding the sample points with maximum expected improve- ment （EI） function, a new Kriging model with higher accuracy is formed. Finally, for verifying accuracy and validity of the Kriging adaptive surrogate model, typical functions and airfoil example are tested in this paper. Test results show that using Kriging adaptive surrogate model, the aerodynamic performance of the airfoil could be efficiently improved.

关 键 词：气动优化 代理模型 自适应选样 差分进化算法 

分 类 号：V211[航空宇航科学与技术—航空宇航推进理论与工程]