郭振东,李存晰,宋立明,李军,丰镇平

• 1:西安交通大学能源与动力工程学院

摘要(Abstract)：

为缩短精细气动形状设计优化所需的最少性能评估次数与任务周期内所能容许的最大性能评估次数之间的差距，基于机器学习领域迁移学习理念，开展了采用知识迁移加速的智能气动设计优化方法研究。首先，搭建了翼型变分自编码器模型，利用其解码器实现了气动形状的智能参数化，同时借助其编码器将已完成任务样本统一至目标任务参数化空间；其次结合单保真度和多保真度代理模型，建立了贝叶斯迁移优化算法；然后，将翼型变分自编码器模型与贝叶斯迁移优化算法相结合，搭建了智能气动形状迁移优化框架；最后，通过任务相关性分析对知识迁移加速优化过程的机理进行了讨论。研究结果表明：通过开展翼型设计优化，智能气动形状迁移优化框架所获得的最优解中位数相较于无知识迁移的变分自编码器优化方法，性能提升了4.8%,比其他各参比方法提升了19.9%以上，验证了该知识迁移策略的有效性。

关键词(KeyWords)： 气动形状设计优化;知识迁移;变分自编码器;贝叶斯优化;多保真度代理模型

Abstract:

Keywords:

基金项目(Foundation): 国家科技重大专项资助项目（2019-Ⅱ-0008-0028）;; 秦创原引用高层次创新创业人才项目（QCYRCXM-2022-210）

作者(Author): 郭振东,李存晰,宋立明,李军,丰镇平

参考文献(References)：

• [1] LI Jichao,ZHANG Mengqi,MARTINS J R R A,et al.Efficient aerodynamic shape optimization with deep-learning-based geometric filtering [J].AIAA Journal,2020,58(10):4243-4259.
• [2] 席光，王志恒，王尚锦.叶轮机械气动优化设计中的近似模型方法及其应用 [J].西安交通大学学报，2007,41(2):125-135,184.XI Guang,WANG Zhiheng,WANG Shangjin.Aerodynamic optimization design of turbomachinery with approximation model method [J].Journal of Xi’an Jiaotong University,2007,41(2):125-135,184.
• [3] 黄松，阳诚武，韩戈，等.采用鲸鱼算法的可控扩散叶型优化设计 [J].西安交通大学学报，2020,54(3):49-57.HUANG Song,YANG Chengwu,HAN Ge,et al.Optimal design of a controlled diffusion airfoil with the whale algorithm [J].Journal of Xi’an Jiaotong University,2020,54(3):49-57.
• [4] 汪祺能，宋立明，郭振东，等.采用动态交互作用分析的叶轮机械优化算法研究 [J/OL].西安交通大学学报，2023(7):1-10 [2022-08-15].http://kns.cnki.net/kcms/detail/61.1069.t.20230110.1409.001.html.WANG Qineng,SONG Liming,GUO Zhendong,et al.Study on turbomachinery optimization based on dynamic interaction analysis [J/OL].Journal of Xi’an Jiaotong University,2023(7):1-10 [2022-08-15].http://kns.cnki.net/kcms/detail/61.1069.t.20230110.1409.001.html.
• [5] 郭振东，宋立明，李军，等.基于子元模型的全局优化与设计空间知识挖掘方法 [J].推进技术，2015,36(2):207-216.GUO Zhendong,SONG Liming,LI Jun,et al.Meta model-based global design optimization and exploration method [J].Journal of Propulsion Technology,2015,36(2):207-216.
• [6] SHAN Songqing,WANG G G.Survey of modeling and optimization strategies to solve high-dimensional design problems with computationally-expensive black-box functions [J].Structural and Multidisciplinary Optimization,2010,41(2):219-241.
• [7] MASTERS D A,TAYLOR N J,RENDALL T C S,et al.Geometric comparison of aerofoil shape parameterization methods [J].AIAA Journal,2017,55(5):1575-1589.
• [8] CHEN Wei,CHIU K,FUGE M D.Airfoil design parameterization and optimization using Bézier generative adversarial networks [J].AIAA Journal,2020,58(11):4723-4735.
• [9] WANG Yuyang,SHIMADA K,FARIMANI A B.Airfoil GAN:encoding and synthesizing airfoils for aerodynamic shape optimization [J/OL].Journal of Computational Design and Engineering,2023:qwad046 [2022-10-26].https://doi.org/10.1093/jcde/qwad046.
• [10] VIANA F A C,SIMPSON T W,BALABANOV V,et al.Special section on multidisciplinary design optimization:metamodeling in multidisciplinary design optimization:how far have we really come [J].AIAA Journal,2014,52(4):670-690.
• [11] 韩忠华.Kriging模型及代理优化算法研究进展 [J].航空学报，2016,37(11):3197-3225.HAN Zhonghua.Kriging surrogate model and its application to design optimization:a review of recent progress [J].Acta Aeronautica et Astronautica Sinica,2016,37(11):3197-3225.
• [12] BAERT L,DUMONT C,BEAUTHIER C,et al.Multidisciplinary design of a low-noise propeller:part II efficient aero-acoustic-mechanical design methodology exploiting surrogate models in an adaptive design space [C]//ASME Turbo Expo 2020:Turbomachinery Technical Conference and Exposition.New York,USA:ASME,2020:16321.
• [13] BHOSEKAR A,IERAPETRITOU M.Advances in surrogate based modeling,feasibility analysis,and optimization:a review [J].Computers & Chemical Engineering,2018,108:250-267.
• [14] LI Kai,KOU Jiaqing,ZHANG Weiwei.Unsteady aerodynamic reduced-order modeling based on machine learning across multiple airfoils [J].Aerospace Science and Technology,2021,119:107173.
• [15] TUCKER P G.Trends in turbomachinery turbulence treatments [J].Progress in Aerospace Sciences,2013,63:1-32.
• [16] MIN A T W,SAGARNA R,GUPTA A,et al.Knowledge transfer through machine learning in aircraft design [J].IEEE Computational Intelligence Magazine,2017,12(4):48-60.
• [17] PAN S J,YANG Qiang.A survey on transfer learning [J].IEEE Transactions on Knowledge and Data Engineering,2010,22(10):1345-1359.
• [18] GUPTA A,ONG Y S,FENG Liang.Insights on transfer optimization:because experience is the best teacher [J].IEEE Transactions on Emerging Topics in Computational Intelligence,2018,2(1):51-64.
• [19] TAN K C,FENG Liang,JIANG Min.Evolutionary transfer optimization:a new frontier in evolutionary computation research [J].IEEE Computational Intelligence Magazine,2021,16(1):22-33.
• [20] VALENZUELA DEL RíO J E,MAVRIS D.Gaussian process surrogate model for levering similar trends across concepts [J].AIAA Journal,2015,53(4):1002-1015.
• [21] WANG Qineng,SONG Liming,GUO Zhendong,et al.Transfer optimization in accelerating the design of turbomachinery cascades [C]//ASME Turbo Expo 2020:Turbomachinery Technical Conference and Exposition.New York,USA:ASME,2020:16321.
• [22] MIN A T W,ALMANDOZ R,GUPTA A,et al.Coping with data scarcity in aircraft engine design [C]//18th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference.Reston,VA,USA:AIAA,2017:4434.
• [23] KINGMA D P,WELLING M.Auto-encoding variational Bayes [C]//2nd International Conference on Learning Representations.New York,USA:ICLR,2014.
• [24] GUO Zhendong,SONG Liming,PARK C,et al.Analysis of dataset selection for multi-fidelity surrogates for a turbine problem [J].Structural and Multidisciplinary Optimization,2018,57(6):2127-2142.
• [25] GUO Zhendong,WANG Qineng,SONG Liming,et al.Parallel multi-fidelity expected improvement method for efficient global optimization [J].Structural and Multidisciplinary Optimization,2021,64(3):1457-1468.
• [26] GUO Zhendong,SUN Wei,SONG Liming,et al.Generative transfer optimization for aerodynamic design [C]//Proceedings of Global Power and Propulsion Society.Zug,Switzerland:GPPS,2021:225.
• [27] GUO Zhendong,ONG Y S,LIU Haitao.Calibrated and recalibrated expected improvements for Bayesian optimization [J].Structural and Multidisciplinary Optimization,2021,64(6):3549-3567.
• [28] DRELA M.XFOIL:an analysis and design system for low Reynolds number airfoils [C]//Low Reynolds Number Aerodynamics.Berlin,Heidelberg:Springer Berlin Heidelberg,1989:1-12.
• [29] CHEN Wei,AHMED F.PaDGAN:learning to generate high-quality novel designs [J].Journal of Mechanical Design,2021,143(3):031703.
• [30] YONEKURA K,MIYAMOTO N,SUZUKI K.Inverse airfoil design method for generating varieties of smooth airfoils using conditional WGAN-gp [J].Structural and Multidisciplinary Optimization,2022,65(6):173.
• [31] SPREITER J R.On the application of transonic similarity rules [EB/OL].[2022-05-05].https://digital.library.unt.edu/ark:/67531/metadc56143/.