基于生成对抗网络的通航飞行器气动布局设计方法

doi:10.7527/S1000-6893.2025.32518

Abstract

Abstract:

Mainstream aerodynamic optimization approaches typically commence with the refinement of initial design stages， concentrating on local optimization within predefined configurations， rather than exploring the diversity of aerodynamic layout alternatives during the conceptual design phase. Recently， generative artificial intelligence has offered a novel solution paradigm for aircraft configuration design. This study presents a generative aerodynamic configuration design methodology based on Generative Adversarial Network （GAN）， with application to a small general aviation aircraft. First， the GAN and its variants are evaluated for their capability in representing parametric space and generating aerodynamic configurations. Based on this analysis， a Conditional Wasserstein GAN with Gradient Penalty （CWGAN-GP） is established for efficient generation of aerodynamic configurations. Second， for wing design under low-speed cruise conditions （Mach number 0.2， angle of attack of 2°）， the representation capability of different generative models in the parameter space and their ability to generate configurations under given conditions are analyzed， demonstrating the advantages of CWGAN-GP in generative design. Finally， for the conceptual aerodynamic configuration design of a general aviation aircraft under typical cruise conditions （Mach number 0.6， angle of attack of 2°）， the CWGAN-GP model successfully generates a variety of aerodynamic configurations using the lift coefficient and lift-to-drag ratio as conditional parameters. By extending beyond the training parameter ranges， the proposed method exhibits strong generalization and extrapolation capabilities， capable of producing different aerodynamic shapes outside the original condition space. These results indicate that the generative design methodology presented offers a promising new paradigm for next-generation aircraft aerodynamic configuration exploration and innovation.

Key words: generative models, aircraft design, aerodynamic configuration design, generative adversarial network, aerodynamics

CLC Number:

V221.3

Youzheng YU, Jiaqing KOU, Weiwei ZHANG. A generative adversarial network based method for aerodynamic configuration design of general aviation aircraft[J]. Acta Aeronautica et Astronautica Sinica, 2026, 47(6): 132518.

Figures/Tables 29

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Table 1

Table 2

Table 3

Table 4

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Table 5

Evaluation metrics of generated results for different target values

指标	CGAN	CWGAN	CWGAN-GP
$B i a s ¯$ /%	0.268	0.057	0.018
$M S E ¯$	3.61×10^-5	5.53×10^-5	1.43×10^-6

Table 5

Fig.11

Fig.12

Fig.13

Table 6

Fig.14

Table 7

Table 8

Table 9

Fig.15

Fig.16

Fig.17

Fig.18

Fig. 19

Fig.20

References 39

[1]	陈小前，姚雯，赵勇，等. 飞行器多学科设计优化理论与应用研究［M］. 2版. 北京：国防工业出版社， 2023： 34.
	CHEN X Q， YAO W， ZHAO Y. Multidisciplinary design optimization of flight vehicles theory and applications［M］. 2nd ed. Beijing： National Defense Industry Press， 2023： 34 （in Chinese）.
[2]	陈树生，贾苜梁，林家豪，等. 生成式模型赋能飞行器技术应用研究进展与展望［J］. 航空学报， 2025， 46（10）： 631194.
	CHEN S S， JIA M L， LIN J H， et al. Empowering aircraft technology applications with generative models： Research progress and prospects［J］. Acta Aeronautica et Astronautica Sinica， 2025， 46（10）： 631194 （in Chinese）.
[3]	DU X S， REN J， LEIFSSON L. Aerodynamic inverse design using multifidelity models and manifold mapping［J］. Aerospace Science and Technology， 2019， 85： 371-385.
[4]	林杰，唐志共，钱炜祺，等. 飞行器生成式气动设计研究进展与展望［J］. 航空学报， 2025， 46（10）： 631679.
	LIN J， TANG Z G， QIAN W Q， et al. Research progress and prospect of aerodynamic configuration design based on generative models for aircraft［J］. Acta Aeronautica et Astronautica Sinica， 2025， 46（10）： 631679 （in Chinese）.
[5]	PETER J E V， DWIGHT R P. Numerical sensitivity analysis for aerodynamic optimization： A survey of approaches［J］. Computers & Fluids， 2010， 39（3）： 373-391.
[6]	LUO J Q， XIONG J T， LIU F. Aerodynamic design optimization by using a continuous adjoint method［J］. Science China Physics， Mechanics & Astronomy， 2014， 57（7）： 1363-1375.
[7]	HOLLAND J H. Adaptation in natural and artificial systems ［M］. Cambridge： The MIT Press； 1992.
[8]	KENNEDY J， EBERHART R. Particle swarm optimization［C］∥Proceedings of ICNN'95 - International Conference on Neural Networks. Piscataway： IEEE Press， 1942-1948.
[9]	RASMUSSEN C E. Gaussian processes in machine learning［M］∥Advanced Lectures on Machine Learning. Berlin， Heidelberg： Springer， 2004： 63-71.
[10]	ZHU L Y， ZHANG W W， KOU J Q， et al. Machine learning methods for turbulence modeling in subsonic flows around airfoils［J］. Physics of Fluids， 2019， 31： 015105.
[11]	张伟伟，寇家庆，刘溢浪，等. 智能赋能流体力学展望［J］. 航空学报， 2021， 42（4）： 524689.
	ZHANG W W， KOU J Q， LIU Y L， et al. Prospect of intelligent empowerment fluid mechanics［J］. Acta Aeronautica et Astronautica Sinica， 2021， 42（4）： 524689 （in Chinese）.
[12]	LI J C， ZHANG M Q， 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.
[13]	LI J C， ZHANG M Q. On deep-learning-based geometric filtering in aerodynamic shape optimization［J］. Aerospace Science and Technology， 2021， 112： 106603.
[14]	ZHAO J C， ZENG L F， SHAO X M. A novel prediction method for unsteady aerodynamic force on three-dimensional folding wing aircraft［J］. Aerospace Science and Technology， 2023， 137： 108287.
[15]	CHEN X Y， LI C N， GONG C L， et al. A study of morphing aircraft on morphing rules along trajectory［J］. Chinese Journal of Aeronautics， 2021， 34（7）： 232-243.
[16]	LIGHTHILL M J. A new method of two-dimensional aerodynamic design：R.&M.No.2112［R］. London： His Majesty’s Stationery Office， 1945.
[17]	刘俊，宋文萍，韩忠华，等. Kriging 模型在翼型反设计中的应用研究［J］. 空气动力学学报， 2014， 32（4）： 518-526.
	LIU J， SONG W P， HAN Z H， et al. Research on application of Kriging model in airfoil reverse design［J］. Acta Aerodynamica Sinica， 2014， 32（4）： 518-526 （in Chinese）.
[18]	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）： 1909-1917.
	HE L， QIAN W Q， LIU T， et al. Inverse design method of airfoil based on deep learning［J］. Journal of Aerospace Power， 2020， 35（9）： 1909-1917 （in Chinese）.
[19]	DU X S， HE P， MARTINS J R R A. Rapid airfoil design optimization via neural networks-based parameterization and surrogate modeling［J］. Aerospace Science and Technology， 2021， 113： 106701.
[20]	DU X S， HE P， MARTINS J R R A. A B-spline-based generative adversarial network model for fast interactive airfoil aerodynamic optimization［C］∥AIAA Scitech 2020 Forum. Reston： AIAA， 2020.
[21]	王景，柳位，谢海润，等. 扩散模型驱动的超临界翼型多目标生成式设计［J］. 航空学报， 2025， 46 （10）： 631210.
	WANG J， LIU W， XIE H R， et al. Multi-objective generative design of supercritical airfoil driven by diffusion model［J］. Acta Aeronautica et Astronautica Sinica， 2025， 46 （10）： 631210 （in Chinese）.
[22]	CHEN Q Y， WANG J， POPE P， et al. Inverse design of two-dimensional airfoils using conditional generative models and surrogate log-likelihoods［J］. Journal of Mechanical Design， 2022， 144（2）： 021712.
[23]	LEI R W， BAI J Q， WANG H， et al. Deep learning based multistage method for inverse design of supercritical airfoil［J］. Aerospace Science and Technology， 2021， 119： 107101.
[24]	楼锦华，陈荣钱，柳家齐，等. 基于门控扩散模型的飞行器气动性能预测与反设计［J］. 航空学报， 2025， 46（10）： 631183.
	LOU J H， CHEN R Q， LIU J Q， et al. Aircraft aerodynamic performance prediction and inverse design based on a gated diffusion model［J］. Acta Aeronautica et Astronautica Sinica， 2025， 46（10）： 631183 （in Chinese）.
[25]	GOODFELLOW I J， POUGET-ABADIE J， MIRZA M， et al. Generative adversarial nets［C］∥Advances in Neural Information Processing Systems 27. Red Hook： Curran Associates， Inc.， 2014： 2672-2680.
[26]	ARJOVSKY M， BOTTOU L. Towards principled methods for training generative adversarial networks ［DB/OL］. arXiv Preprint： 1701.04862， 2017.
[27]	ARJOVSKY M， CHINTALA S， BOTTOU L. Wasserstein generative adversarial networks［C］∥ Proceedings of the 34th International Conference on Machine Learning-Volume 70. New York： ACM， 2017： 214-223.
[28]	MIRZA M， OSINDERO S. Conditional generative adversarial nets［DB/OL］. arXiv Preprint： 1411.1784， 2014.
[29]	MCDONALD R A， GLOUDEMANS J R. Open vehicle sketch pad： An open source parametric geometry and analysis tool for conceptual aircraft design［C］∥AIAA Scitech 2022 Forum. Reston： AIAA， 2022.
[30]	ABRAHAM T， LAZZARA D， HUNSAKER D. Multifidelity comparison of supersonic wave drag prediction methods using axisymmetric bodies［J］. Aerospace， 2024， 11（5）： 359.
[31]	孙宗燕，王强，乔伟，等. 电动飞机分布式螺旋桨—机翼设计分析方法研究［J］. 航空科学技术， 2024（1）： 53-64.
	SUN Z Y， WANG Q， QIAO W， et al. Research on design and analysis method of distributed propeller-wing for electric aircraft［J］. Aeronautical Science & Technology， 2024（1）： 53-64 （in Chinese）.
[32]	ROSAS-CORDOVA J， SANTANA-DELGADO C， HERNANDEZ-ALCANTARA D， et al. Validation of VSPAERO for basic wing simulation［J］. Revista Internacional de Métodos Numéricos Para Cálculo y Diseño En Ingeniería， 2024， 40（4）： 45.
[33]	MAVRIPLIS N， TING K-Y， MOUSTAFA A， et al. Supersonic configurations at low speeds （SCALOS）： Test/simulation correlation studies［C］∥AIAA Scitech 2022 Forum. Reston： AIAA， 2022.
[34]	BELBEN J. Enabling rapid conceptual design using geometry-based multi-fidelity models in VSP［C］∥51st AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition. Reston： AIAA， 2013.
[35]	BRAKE N， MCDONALD R. Physics-based conceptual design flying qualities analysis using OpenVSP and VSPAERO： NASA SBIR Phase I Final Report ［R］. Washington， D.C.： NASA， 2016.
[36]	HEUSEL M， RAMSAUER H， UNTERTHINER T， et al. GANs trained by a two time-scale update rule converge to a local Nash equilibrium［C］∥ Advances in Neural Information Processing Systems 30 （NIPS 2017）. 2017.
[37]	XU Q， HUANG G， YUAN Y， et al. An empirical study on evaluation metrics of generative adversarial networks［DB/OL］. arXiv Preprint： 1806.07755， 2018.
[38]	VAN DER MAATEN L， HINTON G. Visualizing data using t-SNE［J］. Journal of Machine Learning Research， 2008， 9（11）： 2579-2605.
[39]	SEGUI M M， GHAZI G， BOTEZ R M， et al. Design， development and validation of a Cessna citation X aerodynamic model using OpenVSP software［C］∥ 2018 Modeling and Simulation Technologies Conference. Reston： AIAA， 2018.

设计变量	定义	上限	下限
Span/m	半边翼展	10	6
Chord_1/m	翼尖弦长	2	1
Chord_2/m	翼根弦长	5	翼尖弦长长度
Dihedral/（°）	上反角	5	0
Sweep/（°）	后掠角	30	0
t/c	翼型最大厚度	0.2	0.1

网络名称	网络层	神经元个数	激活函数
生成器	隐藏层1	512	ReLU
	隐藏层1	128	ReLU
	隐藏层3	16	ReLU
	输出层	6	Tanh
判别器	隐藏层1	128	LeakyReLU
	隐藏层1	64	LeakyReLU
	隐藏层3	16	LeakyReLU
	输出层	1	Sigmoid （GAN，CGAN）

网络	批次	学习率	迭代	优化器
GAN		0.000 2（G） 0.000 02（D）		Adam
WGAN	128	0.000 05	1 000	RMSprop
WGAN-GP		0.000 1		Adam
CGAN		0.000 2（G） 0.000 02（D）		Adam
CWGAN		0.000 05		RMSprop
CWGAN-GP		0.000 1		Adam

模型	是否引入条件变量	MMD距离	MSE值	MNND
GAN	否	0.193 1
WGAN	否	0.041 5
WGAN-GP	否	0.096
CGAN	是		2×10^-5	0.372 1
CWGAN	是		5.3×10^-5	0.804 8
CWGAN-GP	是		6×10^-6	0.801 7

设计变量	定义	下限	上限	基准
Span_1/m	主翼翼段1展长（半边）	2	4	19.38 （完整）
Span_2/m	主翼翼段2展长（半边）	6	8	19.38 （完整）
Chord_3/m	主翼翼根弦长	3.5	5.5	展弦比 7.8
Chord_1/m	主翼翼尖弦长	0.8	1.5
Chord_2/m	主翼分段处弦长
Area/m²	主翼面积	47	50	48.96
Sweep_1/（°）	主翼翼段1后掠角	0	40
Sweep_2/（°）	主翼翼段2后掠角	0	40
Dihedral_1/（°）	主翼翼段1上反角	0	2	2
Dihedral_2/（°）	主翼翼段2上反角	0	3	2
t/c_1	主翼翼尖翼型最大厚度	0.11	0.24
t/c_2	主翼分段处翼型最大厚度	0.11	0.24
t/c_3	主翼翼根翼型最大厚度	0.11	0.24
X%	主翼前缘相对机身位置	25	45
V_span/m	垂尾展长	3	3.6
V_chord_2/m	垂尾翼根弦长	2	2.7
V_chord_1/m	垂尾翼尖弦长	1.5	2
V_area/m²	垂尾面积	9	11	10.31
V_t/c_2	垂尾翼根翼型最大厚度	0.11	0.24
V_t/c_1	垂尾翼尖翼型最大厚度	0.11	0.24
V_sweep/（°）	垂尾后掠角	30	40
V_x/m	垂尾后缘距离机尾距离	0.1	1.5
H_span/m	平尾展长（半边）	3.5	4.5
H_chord_2/m	平尾翼根弦长	3	4
H_chord_1/m	平尾翼尖弦长	2	3.5
H_area/m²	平尾面积	9	12	11.15
H_t/c_2	平尾翼根翼型最大厚度	0.11	0.24
H_t/c_1	平尾翼尖翼型最大厚度	0.11	0.24
H_sweep/（°）	平尾后掠角	30	45
H_x/%	平尾相对垂尾翼尖位置	10	20
Length/m	机身长度	17	20

A generative adversarial network based method for aerodynamic configuration design of general aviation aircraft

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 29

References 39

Related Articles 15

Recommended Articles

Metrics

Comments

[1]	Zhaoqin PENG, Qicong LI, Haini ZHANG, Hong WU, Yunpeng MA. Improved ACGAN for fault diagnosis based on GRU and convolutional block attention module [J]. Acta Aeronautica et Astronautica Sinica, 2026, 47(2): 431929-431929.
[2]	Pengqian YANG, Yutong CHEN, Junhui LIU, Jiehao YANG, Jiayuan SHAN, Shijun SUN. Aerodynamic and operational characteristics analysis for tandem wing cargo UAV at high angle of attack [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(9): 131056-131056.
[3]	Zhenbing LUO, Hao WANG, Zhijie ZHAO. Theory of dual synthetic jets and its empowerment of advancements in aeronautical technology [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(5): 531821-531821.
[4]	Haixin CHEN, Liangtao FENG, Yuyu DUAN, Yukun GUO, Yue FU, Chenliang PAN, Yishan ZHOU. Aerodynamics-driven monoplane-biplane morphing aircraft [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(5): 531905-531905.
[5]	Xudong LI, Wei ZHONG, Zhen WANG, Tongguang WANG, Jinlong LI. Ratio of propeller thrust to total thrust of ducted propellers [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(4): 230829-230829.
[6]	Junqiang WU, Zhi WEI, Yang LIU, Min ZHONG, Fujun YANG, Yonggang YU. Civil aircraft standard model system and construction of aerodynamic subject database [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(24): 132030-132030.
[7]	Ruixuan XIE, Ziyang ZHANG, Jinye LI, Han BAO, Fanghan LU, Qimin WANG, Dawei WU. Low-boom aerodynamic design and assessment of long-range supersonic passenger aircraft [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(20): 531589-531589.
[8]	Xianchao MENG, Jun TAO. An airfoil inverse design method based on target testing conditional generative adversarial network [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(10): 631182-631182.
[9]	Zhichao WANG, Xinhai CHEN, Liang DENG, Yang LIU, Yufei PANG, Jie LIU. A surface mesh smoothing method for aircraft based on unsupervised learning [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(10): 631172-631172.
[10]	Jie LIN, Zhigong TANG, Weiqi QIAN, Yueqing WANG, Peng ZHANG, Weixia XU, Jie LIU. Research progress and prospects of aircraft aerodynamic design based on generative models [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(10): 631679-631679.
[11]	Dengfeng HU, Yu XIANG, Jun ZHANG, Jiachen YANG, Wenyong WANG. Multi-source data fusion method based on radial basis function generative adversarial network [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(10): 631478-631478.
[12]	Chao AN, Guixi HUO, Yang MENG, Changchuan XIE, Chao YANG. Aerodynamic modeling methods and influence of layout parameters for wingtip⁃hinged multi⁃body combined UAV [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(6): 629587-629587.
[13]	Yanhua ZHANG, Dengcheng ZHANG, Zhangwen ZHOU, Yuchang LEI, Lin LI. Concept and design of virtual rudder surface aircraft based on circulation control: Review [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(6): 629608-629608.
[14]	Chengpeng LIU, Wenping SONG, Changhao GAO, Shaoqiang HAN, Yue WANG, Zhonghua HAN. Air-to-air missile post-stall flip backward launch method [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(20): 129880-129880.
[15]	Zibo ZHUANG, Chunhui ZHANG, Xing CHEN, Jingyuan SHAO, Pakwai CHAN. Clear⁃air turbulence recognition by Doppler⁃wind⁃lidar in terminal area based on DCGAN [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(16): 329748-329748.