Abstract
The use of machine learning in aerodynamic shape optimization problems has significantly increased in recent years. While existing deep learning techniques enable efficient design space exploration on data with an underlying Euclidean or grid-like structure, the direct optimization of non-parametric 3D geometries is still limited. In this article, we propose a geometric deep learning model that generates triangled-based meshed surfaces through the use of a graph variational autoencoder that learns the latent representations of a non-parametric 3D dataset. Once this framework is trained to embed all the input meshes in a properly distributed latent space, its exploration is managed by a genetic algorithm. In this regard, the NSGA-II is the agent in charge of sampling geometries that combine topology and aerodynamic features of the initial ones. Furthermore, in each iteration, it evaluates their aerodynamic performance with CFD in order to guide the optimization process and find the most effective region of the latent space. As a result, those solutions that maximize aerodynamic performance are provided through a Pareto front. The application to a case study and a real-world application is introduced aiming to validate the proposed approach.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Abadía-Heredia R, López-Martín M, Carro B, Arribas J, Pérez J, Le Clainche S (2022) A predictive hybrid reduced order model based on proper orthogonal decomposition combined with deep learning architectures. Expert Syst Appl 187(115):910. https://doi.org/10.1016/j.eswa.2021.115910
Achlioptas P, Diamanti O, Mitliagkas I, Guibas L (2018) Learning representations and generative models for 3D point clouds. In: Proceedings of the 35th international conference on machine learning, pp 40–49, https://doi.org/10.48550/ARXIV.1707.02392
Achour G, Sung W, Pinon O, Mavris D (2020) Development of a conditional generative adversarial network for airfoil shape optimization. In: AIAA SciTech Forum, https://doi.org/10.2514/6.2020-2261
Allen CB, Poole DJ, Rendall TC (2018) Wing aerodynamic optimization using efficient mathematically-extracted modal design variables. Optim Eng 19(2):453–477. https://doi.org/10.1007/s11081-018-9376-7
Asperti A, Trentin M (2020) Balancing reconstruction error and Kullback-Leibler divergence in variational autoencoders. IEEE Access 8:199440–199448. https://doi.org/10.1109/ACCESS.2020.3034828
Bellman R (1957) Dynamic programming. Princeton University Press, New Jersey
Ben-Hamu H, Maron H, Kezurer I, Avineri G, Lipman Y (2018) Multi-chart generative surface modeling. ACM Trans Graph 37(6):1–15. https://doi.org/10.1145/3272127.3275052
Berguin SH, Mavris DN (2014) Dimensionality reduction in aerodynamic design using principal component analysis with gradient information. In: 10th AIAA multidisciplinary design optimization conference, pp 71–87, https://doi.org/10.2514/6.2014-0112
Bishop CM (2006) Pattern recognition and machine learning. Springer, New York
Bouritsas G, Bokhnyak S, Ploumpis S, Bronstein M, Zafeiriou S (2019) Neural 3D Morphable models: spiral convolutional networks for 3D shape representation learning and generation. In: Proceedings of the IEEE/CVF international conference on computer vision, pp 7213–7222, https://doi.org/10.48550/ARXIV.1905.02876
Brock A, Lim T, Ritchie JM, Weston N (2016) Generative and discriminative voxel modeling with convolutional neural networks. arXiv https://doi.org/10.48550/ARXIV.1608.04236
Bronstein MM, Bruna J, LeCun Y, Szlam A, Vandergheynst P (2017) Geometric deep learning: going beyond Euclidean data. IEEE Signal Process Mag 34(4):18–42. https://doi.org/10.1109/MSP.2017.2693418
Bruna J, Zaremba W, Szlam A, LeCun Y (2013) Spectral networks and locally connected networks on graphs. In: ICLR 2014 conference, https://doi.org/10.48550/ARXIV.1312.6203
Cai T, Luo S, Xu K, He D, Liu TY, Wang L (2020) GraphNorm: a principled approach to accelerating graph neural network training. arXiv https://doi.org/10.48550/ARXIV.2009.03294
Chen W, Fuge M (2018) BézierGAN: automatic generation of smooth curves from interpretable low-dimensional parameters. arXiv https://doi.org/10.48550/ARXIV.1808.08871
Chen W, Ramamurthy A (2021) Deep generative model for efficient 3d airfoil parameterization and generation. In: AIAA SciTech Forum, https://doi.org/10.2514/6.2021-1690
Chen W, Chiu K, Fuge M (2019) Aerodynamic design optimization and shape exploration using generative adversarial networks. In: AIAA SciTech Forum, https://doi.org/10.2514/6.2019-2351
Corbera S, Olazagoitia J, Lozano J (2016) Multi-objective global optimization of a butterfly valve using genetic algorithms. ISA Trans 63:401–412. https://doi.org/10.1016/j.isatra.2016.03.008
Deb K (2011) Multi-objective optimization using evolutionary algorithms. John Wiley & Sons, New York
Deb K, Pratap A, Agarwal S, Meyarivan T (2002) A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans Evol Comput 6:182–197. https://doi.org/10.1109/4235.996017
Defferrard M, Bresson X, Vandergheynst P (2016) Convolutional neural networks on graphs with fast localized spectral filtering. In: 30rd conference of NeurIPS, pp 3844–3852, https://doi.org/10.48550/ARXIV.1606.09375
Derksen R, Rogalsky T (2010) Bezier-PARSEC: an optimized aerofoil parameterization for design. Adv Eng Softw 41(7):923–930. https://doi.org/10.1016/j.advengsoft.2010.05.002
Fan H, Su H, Guibas L (2017) A point set generation network for 3D object reconstruction from a single image. in: ieee conference on computer vision and pattern recognition, pp 2463–2471, https://doi.org/10.1109/CVPR.2017.264
Fey M, Lenssen JE (2019) Fast graph representation learning with pytorch geometric. In: ICLR workshop on representation learning on graphs and manifolds, https://doi.org/10.48550/ARXIV.1903.02428
Fey M, Lenssen JE, Weichert F, Müller H (2017) SplineCNN: Fast geometric deep learning with continuous B-spline kernels. In: IEEE/CVF conference on computer vision and pattern recognition, pp 869–877, https://doi.org/10.48550/ARXIV.1711.08920
Foster N, Dulikravich G (1997) Three-dimensional aerodynamic shape optimization using genetic and gradient search algorithms. J Spacecr Rocket 34:36–42. https://doi.org/10.2514/2.3189
Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial nets. Adv Neural Inform Process Syst. https://doi.org/10.48550/ARXIV.1406.2661
Grey ZJ, Constantine PG (2018) Active subspaces of airfoil shape parameterizations. AIAA J 56(5):2003–2017. https://doi.org/10.2514/1.J056054
Guo Y, Wang H, Hu Q, Liu H, Liu L, Bennamoun M (2020) Deep learning for 3D point clouds: a survey. IEEE Trans Pattern Anal Mach Intell 43(12):4338–364. https://doi.org/10.1109/TPAMI.2020.3005434
He X, Li J, Mader CA, Yildirim A, Martins JR (2019) Robust aerodynamic shape optimization: from a circle to an airfoil. Aerosp Sci Technol 87:48–61. https://doi.org/10.1016/j.ast.2019.01.051
Henaff M, Bruna J, LeCun Y (2015) Deep convolutional networks on graph-structured data. arXiv https://doi.org/10.48550/ARXIV.1506.05163
Hou CKJ, Behdinan K (2022) Dimensionality reduction in surrogate modeling: a review of combined methods. Data Sci Eng 7(4):402–427. https://doi.org/10.1007/s41019-022-00193-5
Jabón J, Corbera S, Barea R, Martín-Rabadán J (2023) An evolutive-deformation approach to enhance self-supporting areas in additive manufacturing designs. Comput Ind Eng 182(109):386. https://doi.org/10.1016/j.cie.2023.109386
Kenway GKW, Martins JRRA (2017) Buffet-onset constraint formulation for aerodynamic shape optimization. AIAA J 55(6):1930–1947. https://doi.org/10.2514/1.J055172
Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. In: Proceedings of the 3rd ICLR, https://doi.org/10.48550/ARXIV.1412.6980
Kingma DP, Welling M (2013) Auto-encoding variational bayes. arXiv https://doi.org/10.48550/ARXIV.1312.6114
Kolotouros N, Pavlakos G, Daniilidis K (2019) Convolutional mesh regression for single-image human shape reconstruction. In: IEEE/CVF conference on computer vision and pattern recognition, pp 4496–4505, https://doi.org/10.1109/CVPR.2019.00463
Kulfan B (2008) Universal parametric geometry representation method. J Aircr 45:142–1587. https://doi.org/10.2514/1.29958
Kullback S, Leibler RA (1951) On information and sufficiency. Ann Math Stat 22(1):79–86. https://doi.org/10.1214/aoms/1177729694
Lepine J, Guibault F, Trepanier JY, Pepin F (2001) Optimized nonuniform rational B-spline geometrical representation for aerodynamic design of wings. AIAA J 39(11):2033–2041. https://doi.org/10.2514/2.1206
Li J, Zhang M (2021) On deep-learning-based geometric filtering in aerodynamic shape optimization. Aerosp Sci Technol 112(106):603. https://doi.org/10.1016/j.ast.2021.106603
Li J, Zhang M, Martins J, Shu C (2020) Efficient aerodynamic shape optimization with deep-learning-based geometric filtering. AIAA J 58:1–17. https://doi.org/10.2514/1.J059254
Li J, Du X, Martins JR (2022) Machine learning in aerodynamic shape optimization. Prog Aerosp Sci 134(100):849. https://doi.org/10.1016/j.paerosci.2022.100849
Litany O, Bronstein A, Bronstein M, Makadia A (2018) Deformable shape completion with graph convolutional autoencoders. In: IEEE/CVF conference on computer vision and pattern recognition, pp 1886–1895, https://doi.org/10.1109/CVPR.2018.00202
Loshchilov I, Hutter F (2018) Decoupled weight decay regularization. In: ICLR 2019 conference on blind submission, https://doi.org/10.48550/ARXIV.1711.05101
Lyu P, Kenway G, Martins J (2015) Aerodynamic shape optimization investigations of the common research model wing benchmark. AIAA J 53(4):968–985. https://doi.org/10.2514/1.J053318
Masters DA, Taylor NJ, Rendall TCS, Allen CB, Poole DJ (2017) Geometric comparison of aerofoil shape parameterization methods. AIAA J 55(5):1575–1589. https://doi.org/10.2514/1.J054943
Maturana D, Scherer S (2015) VoxNet: a 3D convolutional neural network for real-time object recognition. In: IEEE/RSJ international conference on intelligent robots and systems, pp 922–928, https://doi.org/10.1109/IROS.2015.7353481
Moschoglou S, Ploumpis S, Nicolaou M, Papaioannou A, Zafeiriou S (2020) 3DFaceGAN: adversarial nets for 3D face representation, generation, and translation. Int J Comput Vision 128:2534–2551. https://doi.org/10.1007/s11263-020-01329-8
Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L, et al. (2019) Pytorch: an imperative style, high-performance deep learning library. In: 33rd conference of NeurIPS, pp 8024–8035, https://doi.org/10.48550/ARXIV.1912.01703
Piegl L, Tiller W (1997) The NURBS book. Springer, New York
Poole DJ, Allen CB, Rendall TC (2019) Efficient Aero-structural wing optimization using compact aerofoil decomposition. In: AIAA Scitech Forum, https://doi.org/10.2514/6.2019-1701
Qi CR, Su H, Mo K, Guibas LJ (2017) PointNet: deep learning on point sets for 3D classification and segmentation. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 652–660, https://doi.org/10.48550/ARXIV.1612.00593
Qiu Y, Bai J, Liu N, Wang C (2018) Global aerodynamic design optimization based on data dimensionality reduction. Chin J Aeronaut 31(4):643–659. https://doi.org/10.1016/j.cja.2018.02.005
Rajnarayan DG, Ning A, Mehr JA (2018) Universal airfoil parametrization using B-splines. In: AIAA multidisciplinary analysis and optimization conference, https://doi.org/10.2514/6.2018-3949
Ranjan A, Bolkart T, Sanyal S, Black MJ (2018) Generating 3D faces using convolutional mesh autoencoders. In: Computer vision EECV, pp 725–741, https://doi.org/10.1007/978-3-030-01219-9_43
Ravi N, Reizenstein J, Novotny D, Gordon T, Lo WY, Johnson J, Gkioxari G (2020) Accelerating 3D deep learning with PyTorch3D. In: NeurIPS WiML workshop, https://doi.org/10.48550/ARXIV.2007.08501
Rogalsky T, Derksen RW, Kocabiyik S (1999) Differential evolution in aerodynamic optimization. Can Aeronaut Space J 46(4):183–190
Secanell M, Gamboa P, Suleman A (2006) Design of a morphing airfoil using aerodynamic shape optimization. AIAA J 44:1550–1562. https://doi.org/10.2514/1.18109
Sigmund O, Maute K (2013) Topology optimization approaches. Struct Multidisc Optim 48:1031–1055. https://doi.org/10.1007/s00158-013-0978-6
Sobieczky H (1999) parametric airfoils and wings. In: Notes on numerical fluid mechanics, pp 71–87, https://doi.org/10.1007/978-3-322-89952-1_4
Tan Q, Gao L, Lai YK, Xia S (2018a) Variational autoencoders for deforming 3D mesh models. In: IEEE/CVF conference on computer vision and pattern recognition, pp 5841–5850, https://doi.org/10.1109/CVPR.2018.00612
Tan Q, Gao L, Lai YK, Yang J, Xia S (2018b) Mesh-based autoencoders for localized deformation component analysis. Proc AAAI Conf Artif Intell 32(1):2452–2459. https://doi.org/10.1609/aaai.v32i1.11870
Tretschk E, Tewari A, Zollhofer M, Golyanik V, Theobalt C (2020) DEMEA: deep mesh autoencoders for non-rigidly deforming objects. In: Computer vision EECV, pp 601–617, https://doi.org/10.1007/978-3-030-58548-8_35
Viswanath A, Forrester AIJ, Keane AJ (2011) Dimension reduction for aerodynamic design optimization. AIAA J 49(6):1256–1266. https://doi.org/10.2514/1.J050717
Viswanath A, Forrester AIJ, Keane AJ (2014) Constrained design optimization using generative topographic mapping. Chin J Aeronaut 52(5):1010–1023. https://doi.org/10.2514/1.J052414
Wang N, Zhang Y, Li Z, Fu Y, Liu W, Jiang YG (2018) Pixel2Mesh: generating 3D mesh models from single RGB images. In: Computer vision EECV, pp 55–71, https://doi.org/10.1007/978-3-030-01252-6_4
Wang W, Ceylan D, Mech R, Neumann U (2019) 3DN: 3D deformation network. In: IEEE/CVF conference on computer vision and pattern recognition, pp 1038–1046, https://doi.org/10.1109/CVPR.2019.00113
Wu J, Zhang C, Xue T, Freeman B, Tenenbaum J (2016) Learning a probabilistic latent space of object shapes via 3D generative-adversarial modeling. Adv Neural Inform Process Syst. https://doi.org/10.48550/ARXIV.1610.07584
Yan X, Zhu J, Kuang M, Wang X (2019) Aerodynamic shape optimization using a novel optimizer based on machine learning techniques. Aerosp Sci Technol 86:826–835. https://doi.org/10.1016/j.ast.2019.02.003
Yuan YJ, Lai YK, Yang J, Duan Q, Fu H, Gao L (2020) Mesh variational autoencoders with edge contraction pooling. In: IEEE/CVF Conference on computer vision and pattern recognition workshops, pp 1105–1112, https://doi.org/10.1109/CVPRW50498.2020.00145
Zhang A, Lipton ZC, Li M, Smola AJ (2021) Dive into deep learning. arXiv https://doi.org/10.48550/ARXIV.2106.11342
Acknowledgements
The authors would like to thank Meta and Linux Foundation for creating, publicly sharing, and maintaining the open-source Pytorch and Pytorch Geometric frameworks for AI collaborative research. Furthermore, the authors also acknowledge the internal support of Nebrija University to carry out the PhD program in Industrial and Computer Science Technology (D019).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Replication of results
The results presented in this paper were obtained using a Python implementation of the proposed methodology and algorithms. The code to reproduce the case study is available through GitHub (https://github.com/JJab0n). Researchers or interested parties are welcome to contact the authors for further explanations.
Additional information
Responsible editor: Gang Li
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Jabón, J., Corbera, S., Álvarez, R. et al. Aerodynamic shape optimization using graph variational autoencoders and genetic algorithms. Struct Multidisc Optim 67, 32 (2024). https://doi.org/10.1007/s00158-024-03771-5
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00158-024-03771-5