Adaptive modeling strategy for constrained global optimization with application to aerodynamic wing design

被引:79
作者
Bartoli, N. [1 ]
Lefebvre, T. [1 ]
Dubreuil, S. [1 ]
Olivanti, R. [2 ]
Priem, R. [1 ]
Bons, N. [3 ]
Martins, J. R. R. A. [3 ]
Morlier, J. [4 ]
机构
[1] Univ Toulouse, ONERA DTIS, Toulouse, France
[2] Univ Toulouse, ISAE SUPAERO, Toulouse, France
[3] Univ Michigan, Ann Arbor, MI 48109 USA
[4] Univ Toulouse, ISAE SUPAERO INSA Mines Albi UPS, CNRS UMR5312, Inst Clement Ader, Toulouse, France
基金
欧盟地平线“2020”;
关键词
Surrogate modeling; Global optimization; Multimodal optimization; Mixture of experts; Aerodynamic shape optimization; Wing design; INFILL SAMPLING CRITERIA; SHAPE OPTIMIZATION; EVOLUTION STRATEGY; ALGORITHM; APPROXIMATION; ADJOINT;
D O I
10.1016/j.ast.2019.03.041
中图分类号
V [航空、航天];
学科分类号
08 ; 0825 ;
摘要
Surrogate models are often used to reduce the cost of design optimization problems that involve computationally costly models, such as computational fluid dynamics simulations. However, the number of evaluations required by surrogate models usually scales poorly with the number of design variables, and there is a need for both better constraint formulations and multimodal function handling. To address this issue, we developed a surrogate-based gradient-free optimization algorithm that can handle cases where the function evaluations are expensive, the computational budget is limited, the functions are multimodal, and the optimization problem includes nonlinear equality or inequality constraints. The proposed algorithm-super efficient global optimization coupled with mixture of experts (SEGOMOE)-can tackle complex constrained design optimization problems through the use of an enrichment strategy based on a mixture of experts coupled with adaptive surrogate models. The performance of this approach was evaluated for analytic constrained and unconstrained problems, as well as for a multimodal aerodynamic shape optimization problem with 17 design variables and an equality constraint. Our results showed that the method is efficient and that the optimum is much less dependent on the starting point than the conventional gradient-based optimization. (C) 2019 Elsevier Masson SAS. All rights reserved.
引用
收藏
页码:85 / 102
页数:18
相关论文
共 88 条
[1]  
Allmaras S, 1992, 30 AER SCI M EXH AER
[2]  
ANJOS M., 2009, Mopta 2008 benchmark
[3]  
[Anonymous], SAND2009SAND6265 SAN
[4]  
[Anonymous], 2013, 21 AIAA COMP FLUID D
[5]  
[Anonymous], 2016, 17 AIAA ISSMO MULT A
[6]  
[Anonymous], 2015, THESIS
[7]  
[Anonymous], 2009, THE NOMAD PROJECT
[8]  
[Anonymous], STRUCT MULTIDISCIPL
[9]  
[Anonymous], 269 AIAA
[10]  
[Anonymous], AIAA PAPER