An Airfoil Inverse Design Method Based on Target Testing Conditional Generation Adversarial Network

MENG Xian-Chao; TAO Jun

doi:10.7527/S1000-6893.2024.31182

ACTA AERONAUTICAET ASTRONAUTICA SINICA >

0 0 - 0

DOI: https://doi.org/10.7527/S1000-6893.2024.31182

An Airfoil Inverse Design Method Based on Target Testing Conditional Generation Adversarial Network

MENG Xian-Chao ,
TAO Jun

Expand

Received date: 2024-09-10

Revised date: 2024-11-14

Online published: 2024-11-18

Fold

Abstract

In this study, a Target Testing Conditional Generative Adversarial Network (TT-CGAN) is developed and applied to airfoil inverse design. This network extends the Conditional Generative Adversarial Network (CGAN) by integrating a Multi-Layer Perceptron (MLP) checker, enhancing the capability of CGAN in evaluating the impact of additional conditions on the target testing. Utilizing the UIUC airfoil database, 797 real airfoils were selected, and their corresponding aerodynamic parameters were calculated using the Reynolds-Averaged Navier-Stokes (RANS) method to construct a comprehensive airfoil database. The airfoil shapes were parameterized using the Class Shape. Transformation (CST) method, reducing the geometric parameter dimensionality from 100 to 14 CST parameters. Multi-source aerodynamic parameters, including lift coefficient, drag coefficient, and surface pressure distribution, were fused using a feature-level fusion approach. These fused parameters were compared with aerodynamic parameters based solely on lift and drag coefficients, which served as auxiliary conditions for the network during the airfoil inverse design process. The results indicate that the TT-CGAN-based inverse design method generates more accurate airfoils, with an average mean square error of 0.1779% and an average absolute error of 0.1351% in airfoil geometry. The generated airfoils were further validated through numerical simulations using the RANS method, demonstrating an average relative error of 3.5998% for the lift coefficient and 3.7239% for the drag coefficient, confirming that the generated airfoils meet the specified design criteria. Analysis of the lift-to-drag ratio distributions revealed that 40% of the test airfoils achieved lift-to-drag ratios within the [20, 30) range, compared to only 16% in the training set. This finding highlights the method’s capability to make accurate predictions even within data-sparse regions, showcasing its generalizability.

Key words： airfoil inverse design; CGAN; multi-model data; class shape transformation; parameterization

Cite this article

MENG Xian-Chao , TAO Jun . An Airfoil Inverse Design Method Based on Target Testing Conditional Generation Adversarial Network[J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 0 : 0 -0 . DOI: 10.7527/S1000-6893.2024.31182

References

[1] 卜月鹏, 宋文萍, 韩忠华,等. 基于 CST 参数化方法的翼型气动优化设计 [C]// 全国计算流体力学会议. 中国空气动力学会;中国力学学会;中国航空学会;中国宇航学会, 2012.
[2] 王清，招启军. 基于遗传算法的旋翼翼型综合气动优化设计[J]. 航空动力学报，2016, 31(6): 1486-1495. WANG Qing, ZHAO Qi-jun. Synthetical optimization design of rotor airfoil by genetic algorithm[J]. Journal of Aerospace Power, 2016, 31(6): 1486-1495.(in Chinese).
[3] 韩忠华, 许晨舟, 乔建领,等. 基于代理模型的高效全局气动优化设计方法研究进展[J]. 航空学报, 2020, 41(5):41.
[4] 白俊强, 雷锐午, 杨体浩,等. 基于伴随理论的大型客机气动优化设计研究进展[J]. 航空学报, 2019, 40(1):18.
[5] 夏露, 常彦鑫, 张龙. 改进的模拟退火算法在翼型设计中的应用[J]. 飞行力学, 2008, 26(1):4.
[6] 许瑞飞, 邓一菊, 钱瑞战. 气动优化设计及其对 CFD的需求[J]. 航空科学技术, 2011(2):3.
[7] 穆雪峰. 多学科设计优化代理模型技术的研究和应用[D]. 南京航空航天大学, 2004.
[8] 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):29.
[9] 郭丽丽,丁世飞.深度学习研究进展[J].计算机科学,2015,42(05):28-33.
[10] 廖鹏,姚磊江,白国栋,等.基于深度学习的混合翼型前缘压力分布预测[J].航空动力学报, 2019(8):8.DOI:10.13224/j.cnki.jasp.2019.08.014.
[11] 陈海,钱炜祺,何磊.基于深度学习的翼型气动系数预测[J].空气动力学学报,2018,36(02):294-299.
[12] Jun T ,Gang S .An artificial neural network approach for aerodynamic performance retention in airframe noise reduction design of a 3D swept wing model[J].Chinese Journal of Aeronautics,2016,29(05):1213-1225.
[13] 王超杰,何磊,李川,等.基于注意力机制的翼型反设计方法[J].航空动力学报, 2024, 39:20230106.DOI:10.13224/j.cnki.jasp.20230106.
[14] 赵国庆，招启军. 基于目标压力分布的旋翼先进气动外形反设计分析方法[J]. 航空学报, 2014, 35(3): 744 755. ZHAO G Q, ZHAO Q J. Inverse design analysis method on rotor with advanced aerodynamic configuration based upon target pressure distribution[J]. Acta Aeronautica et Astronautica Sinica, 2014, 35(3):744-755.(in Chinese)
[15] Lighthill M J. A new method of two-dimensional aerodynamic design[J]. 1945.
[16] Steger J L, Klineberg J M. A finite-difference method for transonic airfoil design[J]. AIAA Journal, 1973, 11(5): 628-635.
[17] Carlson L A. Transonic airfoil analysis and design using Cartesian coordinates[J]. Journal of Aircraft, 1976, 13(5): 349-356.
[18] Sobieczky H, Yu N J, Fung K Y, et al. New method for designing shock-free transonic configurations[J]. AIAA Journal, 1979, 17(7): 722-729.
[19] Takanashi S. Iterative three-dimensional transonic wing design using integral equations[J]. Journal of aircraft, 1985, 22(8): 655-660.
[20] Henne P A. Inverse transonic wing design method[J]. Journal of Aircraft, 1981, 18(2): 121-127.
[21] CAMPBELL R, SMITH L. A hybrid algorithm for transonic airfoil and wing design[C]//5th Applied Aerodynamics Conference. 1987: 2552.
[22] 华俊, 张仲寅, 乔志德, 等. 一种跨声速翼型设计方法及设计诸例[J]. 空气动力学学报, 1990, 8(2): 117-123.
[23] 白俊强, 华俊, 张仲寅. 基于欧拉方程的跨声速翼型设计[J]. 空气动力学学报, 1997, 15(4): 458-461.
[24] 詹浩, 华俊, 张仲寅. 基于余量修正原理的多翼面气动力反设计方法[J]. 航空学报, 2003, 24(5): 411-413.
[25] Bui-Thanh T, Damodaran M, Willcox K. Aerodynamic data reconstruction and inverse design using proper orthogonal decomposition[J]. AIAA journal, 2004, 42(8): 1505-1516.
[26] 白俊强,邱亚松,华俊.改进型Gappy POD翼型反设计方法[J].航空学报, 2013, 34(4).DOI:10.7527/S1000-6893.2013.0135.
[27] 宋超,刘红阳,周铸,等.基于生成拓扑映射的气动外形反设计方法研究[J].西北工业大学学报, 2022(004):040.
[28] 吴秋雨.基于生成式对抗网络的气动外形优化方法研究[J].[2024-07-30].
[29] Sekar V , Zhang M , Shu C ,et al. Inverse Design of Airfoil Using a Deep Convolutional Neural Network[J].AIAA Journal, 2019, 57(3):1-11.DOI:10.2514/1.J057894.
[30] 何磊,钱炜祺,刘滔,等.基于深度学习的翼型反设计方法[J].航空动力学报, 2020, 35(9):9.DOI:10.13224/j.cnki.jasp.2020.09.013.
[31] 黄礼铿, 高正红, 张德虎. 基于变可信度代理模型的气动优化[J]. 空气动力学学报, 2013, 31(6):783-788.）
[32] 刘璟,边枭,徐冠峰,李书,王建林.基于多可信度代理模型的尾喷管优化设计[J/OL].航空工程进展:1-11[2022-05-16]
[33] 张立, 陈江涛, 熊芬芬,等. 基于元学习的多可信度深度神经网络代理模型[J]. 机械工程学报, 2022, 58(1):11.
[34] 寇家庆,张伟伟,叶正寅.基于分层思路的动态非线性气动力建模方法优先出版[C]//中国空气动力学会;中国力学学会.中国空气动力学会;中国力学学会, 2015:13.DOI:JournalArticle/5b3b5acbc095d70f006e8155.
[35] Gao J, Li P, Chen Z, et al. A survey on deep learning for multimodal data fusion[J]. Neural Computation, 2020, 32(5): 829-864.
[36] 唐志共,朱林阳,向星皓等.智能空气动力学若干研究进展及展望[J].空气动力学学报,2023,41(07):1-35.
[37] 田洁华，孙迪，屈峰，等．基于CST?GAN的翼型参数化方法［J］．航空学报，2023，44（18）：128280．
[38] TIAN J H, SUN D, QUF, et al. Airfoil parameterization method based on CST-GAN［J］. Acta Aeronautica et Astronautica Sinica,2023,44(18):128280(in Chinese).doi:10.7527/S1000-6893.2023.28280.
[39] Reynolds O. On the dynamic theory of incompressible viscous flows and determination of the criterion J .Philosophical Transactions of the Royal Society of London A,1894,186:123-161
[40] Wilcox D C. Turbulence Modeling for CFD. Anaheim: DCW Industries[J]. 1998.
[41] Sagaut P. Large eddy simulation of incompressible flows[M]3rd ed. Berlin: Springer,2006.
[42] Yan C, Yu J J, Li J Z. Scheme effect and grid dependency in CFD computations of heat transfer[J]. Acta Aerodynamica Sinica, 2006, 24(1): 125-130.
[43] Jiang G S, Shu C W. Efficient implementation of weighted ENO schemes[J]. Journal of computational physics, 1996, 126(1): 202-228.
[44] Jameson A, Yoon S. Lower-upper implicit schemes with multiple grids for the Euler equations[J]. AIAA journal, 1987, 25(7): 929-935.
[45] Wright M J, Candler G V, Prampolini M. Data-parallel lower-upper relaxation method for the Navier-Stokes equations[J]. AIAA journal, 1996, 34(7): 1371-1377.
[46] Spalart P, Allmaras S. A one-equation turbulence model for aerodynamic flows[C]//30th aerospace sciences meeting and exhibit. 1992: 439.
[47] Kulfan B, Bussoletti J. " Fundamental" Parameteric Geometry Representations for Aircraft Component Shapes[C]//11th AIAA/ISSMO multidisciplinary analysis and optimization conference. 2006: 6948.
[48] Kulfan B M. Recent extensions and applications of the ‘CST’universal parametric geometry representation method[J]. The Aeronautical Journal, 2010, 114(1153): 157-176.
[49] Kulfan B. A Universal Parametric Geometry Representation Method-" CST"[C]//45th AIAA Aerospace Sciences Meeting and Exhibit. 62.
[50] 徐亚峰, 刘学军, 吕宏强. 基于CST参数化方法的翼型快速设计[J]. 航空计算技术, 2011, 41(5):7.
[51] 关晓辉, 李占科, 宋笔锋. CST气动外形参数化方法研究[J]. 航空学报, 2012(04):625-633.
[52] Mirza M, Osindero S. Conditional generative adversarial nets[J]. arXiv preprint arXiv:1411.1784, 2014.
[53] 张虎成,李雷孝,刘东江.多模态数据融合研究综述[J/OL].计算机科学与探索,1-22[2024-07-30].

Options

Outlines

模态框（Modal）标题

Abstract

Cite this article

References