A comparative study of learning techniques for the compressible aerodynamics over a transonic RAE2822 airfoil. (English) Zbl 1521.76279
Summary: In this study, the modeling of the compressible pressure field on the RAE 2822 airfoil using deep learning (DL) is investigated. The objective is to generate, at low cost, the complete Mach envelope from a given aerodynamic database. A dataset with more than 2.000 RANS simulations at various angles of attack and Mach numbers has been created, and validated, to train fast and accurate surrogate models model of the pressure field. The procedure is inspired by the well-known analytical Prandtl-Glauert transformation, which maps solutions from one Mach number to another, but however fails in the transonic regime because of the strong non-linearities due to shock waves. Thus, instead of a classical direct generation of the pressure fields, DL is used here to perform a geometric transformation from an incompressible pressure field towards a compressible solution. The key question addressed in this study is: how accurate and sample efficient are DL techniques compared with classical surrogate modeling tools? To answer this question, two types of DL are investigated (CNN and GCNN) and compared with Proper Orthogonal Decomposition (POD) coupled to a Gaussian Process Regression (GPR). These methods are trained on various input resolutions and database sizes. Results show that DL models are able to accurately predict the intensity and location of the shock over a wide range of angles of attack and Mach numbers up to the transonic regime. Predictions of the lift coefficient reveal a relative error lower than 2% with respect to the high-fidelity data. Because of their inherent nonlinear nature, CNN and GCNN provide more accurate results even for small training datasets, whereas POD encounters difficulties to reconstruct properly shock waves. Since CNN requires inputs and outputs in a pixel-like format, it suffers from interpolation errors that can only be mitigated with fine resolutions, making the training more difficult. By performing convolutions directly on unstructured data, GCNN eliminates this interpolation error and provides the best accuracy and sample efficiency. This study shows that advanced DL techniques, such as CNN and GCNN, are capable of predicting complex flows, and outperform classical tools on such tasks.
MSC:
| 76G25 | General aerodynamics and subsonic flows |
References:
| [1] | Karman, T.v., Compressibility effects in aerodynamics, J Aeronaut Sci, 8, 9, 337-356 (1941) · JFM 67.0853.01 |
| [2] | Glauert, H., The effect of compressibility on the lift of an aerofoil, Proc R Soc Lond Ser A, 118, 779, 113-119 (1928) · JFM 54.0919.02 |
| [3] | Tsien, H., Two-dimensional subsonic flow of compressible fluids, J Aeronaut Sci, 6, 10, 399-407 (1939) · Zbl 0063.07865 |
| [4] | Zhu, J.; Liu, L.; Liu, T.; Shi, Y.; Su, W.; Wu, J., Lift and drag in two-dimensional steady viscous and compressible flow: I. Far-field formulae analysis and numerical confirmation, (AIAA AVIATION FORUM, 45th AIAA fluid dynamics conference (2015)) |
| [5] | Brion, V.; Dandois, J.; Abart, J.-C.; Paillart, P., Experimental analysis of the shock dynamics on a transonic laminar airfoil, Progr Flight Phys, 9, 365-386 (2017) |
| [6] | Dandois, J., Experimental study of transonic buffet phenomenon on a 3D swept wing, Phys Fluids, 28, 1, Article 016101 pp. (2016) |
| [7] | Liu, Z.; Zha, G., Transonic airfoil performance enhancement using co-flow jet active flow control, (AIAA 2016-3472 (2016)) |
| [8] | Liu, J.; Song, W.; Han, Z.; Zhang, Y., Efficient aerodynamic shape optimization of transonic wings using a parallel infilling strategy and surrogate models, Struct Multidiscip Optim, 55, 925-943 (2016) |
| [9] | Bourguet, R.; Braza, M.; Dervieux, A., Reduced-order modeling for unsteady transonic flows around an airfoil, Phys Fluids, 19, 11, Article 111701 pp. (2007) · Zbl 1182.76077 |
| [10] | Opgenoord, M.; Drela, M.; Willcox, K., Influence of transonic flutter on the conceptual design of next-generation transport aircraft, AIAA J, 57, 5, 1973-1987 (2019) |
| [11] | Renganathan, S. A.; Maulik, R.; Rao, V., Machine-learning for nonintrusive model order reduction of the parametric inviscid transonic flow past an airfoil (2019), arXiv:1911.07943 |
| [12] | Bertrand, X.; Tost, F.; Champagneux, S., Wing airfoil pressure calibration with deep learning, (AIAA 2019-3066 (2019)) |
| [13] | Duru, C.; Alemdar, H.; Baran, O., A deep learning approach for the transonic flow field predictions around airfoils, Comput & Fluids, 236, Article 105312 pp. (2022) · Zbl 1521.76290 |
| [14] | Greenman, R.; Roth, K., High-lift optimization design using neural networks on a multi-element airfoil, J Fluids Eng, 121, 2, 434-440 (1999) |
| [15] | Tompson, J.; Schlachter, K.; Sprechmann, P.; Perlin, K., Accelerating eulerian fluid simulation with convolutional networks, (Proceedings of the 34th international conference on machine learning-volume 70 (2017), JMLR. org), 3424-3433 |
| [16] | Illarramendi, E.; Alguacil, A.; Bauerheim, M.; Misdariis, A.; Benedicte, C.; Benazera, E., Towards an hybrid computational strategy based on deep learning for incompressible flows, (AIAA 2020-3058 (2020)) |
| [17] | Ajuria Illarramendi, E.; Bauerheim, M.; Cuenot, B., Performance and accuracy assessments of an incompressible fluid solver coupled with a deep convolutional neural network, Data-Centric Eng, 3, Article e2 pp. (2022) |
| [18] | Raissi, M.; Karniadakis, G. E., Hidden physics models: machine learning of nonlinear partial differential equations, J Comput Phys, 357, 125-141 (2018) · Zbl 1381.68248 |
| [19] | Alguacil, A.; Bauerheim, M.; Jacob, M. C.; Moreau, S., Predicting the propagation of acoustic waves using deep convolutional neural networks, (AIAA 2020-2513 (2020)) |
| [20] | Wang, R.; Kashinath, K.; Mustafa, M.; Albert, A.; Yu, R., Towards physics-informed deep learning for turbulent flow prediction (2019), arXiv:1911.08655 |
| [21] | Lapeyre, C.; Misdariis, A.; Cazard, N.; Veynante, D.; Poinsot, T., Training convolutional neural networks to estimate turbulent sub-grid scale reaction rates, Combust Flame, 203, 255-264 (2019) |
| [22] | Krügener, M.; Zapata Usandivaras, J. F.; Bauerheim, M.; Urbano, A., Coaxial-injector surrogate modeling based on Reynolds-averaged Navier-Stokes simulations using deep learning, J Propuls Power, 38, 5, 783-798 (2022) |
| [23] | Zapata Usandivaras, J. F.; Urbano, A.; Bauerheim, M.; Cuenot, B., Data driven models for the design of rocket injector elements, Aerospace, 9, 10 (2022) |
| [24] | Miyanawala, T. P.; Jaiman, R. K., A hybrid data-driven deep learning technique for fluid-structure interaction (2018), arXiv:1807.09591 |
| [25] | Wu, H.; Liu, X.; An, W.; Chen, S.; Lyu, H., A deep learning approach for efficiently and accurately evaluating the flow field of supercritical airfoils, Comput & Fluids, 198, Article 104393 pp. (2020) · Zbl 1519.76017 |
| [26] | Baqué, P.; Remelli, E.; Fleuret, F.; Fua, P., Geodesic convolutional shape optimization (2018), CoRR arXiv:1802.04016 |
| [27] | Thuerey, N.; Weissenow, K.; Mehrotra, H.; Mainali, N.; Prantl, L.; Hu, X., Well, how accurate is it? A study of deep learning methods for Reynolds-averaged Navier-Stokes simulations (2018), CoRR arXiv:1810.08217 |
| [28] | Bhatnagar, S.; Afshar, Y.; Pan, S.; Duraisamy, K.; Kaushik, S., Prediction of aerodynamic flow fields using convolutional neural networks, Comput Mech, 64, 2, 525-545 (2019) · Zbl 1468.76051 |
| [29] | Goodfellow, I.; Bengio, Y.; Courville, A., Deep learning (2016), MIT Press, http://www.deeplearningbook.org · Zbl 1373.68009 |
| [30] | Mathieu, M.; Couprie, C.; LeCun, Y., Deep multi-scale video prediction beyond mean square error (2015), CoRR arXiv:1511.05440 |
| [31] | Chapin, V.; Carlan, N.; Heppel, P., Performance optimization of interacting sails though fluid structure coupling, Int J Small Craft Technol, 153, 103-116 (2011) |
| [32] | Viola, I.; Chapin, V.; Speranza, N.; Biancolini, M., Optimal airfoil’s shapes by high fidelity CFD, Aircr Eng Aerosp Technol, 90, 1000-1011 (2018) |
| [33] | Gourdain, N.; Dewandel, K., Quantification of the numerical solution sensitivity to unknown parameters: Application to compressor and turbine flows, (8 (2012)) |
| [34] | Spalart, P.; Allmaras, S., A one-equation turbulence model for aerodynamic flows, AIAA, 439 (1992) |
| [35] | Cook, P.; Firmin, M.; McDonald, M., Aerofoil RAE 2822: Pressure distributions, and boundary layer and wake measurements (1977), RAE |
| [36] | Morimoto, M.; Fukami, K.; Zhang, K.; Nair, A. G.; Fukagata, K., Convolutional neural networks for fluid flow analysis: toward effective metamodeling and low dimensionalization, Theor Comput Fluid Dyn, 35, 633-658 (2021) |
| [37] | Kingma, D.; Ba, J., Adam: A method for stochastic optimization (2014), arXiv:1412.6980 |
| [38] | Ronneberger, O.; Fischer, P.; Brox, T., U-net: Convolutional networks for biomedical image segmentation (2015), arXiv:1505.04597 |
| [39] | Yu, F.; Koltun, V., Multi-scale context aggregation by dilated convolutions (2015), URL https://arxiv.org/abs/1511.07122 |
| [40] | Burkardt, J.; Gunzburger, M.; Lee, H.-C., POD and CVT-based reduced-order modeling of Navier-Stokes flows, Comput Methods Appl Mech Engrg, 196, 1-3, 337-355 (2006) · Zbl 1120.76323 |
| [41] | Legresley P, Alonso J. Airfoil design optimization using reduced order models based on proper orthogonal decomposition. In: Fluids 2000 conference and exhibit. 2000, http://dx.doi.org/10.2514/6.2000-2545. |
| [42] | Berkooz, G.; Holmes, P.; Lumley, J. L., The proper orthogonal decomposition in the analysis of turbulent flows, Annu Rev Fluid Mech, 25, 1, 539-575 (1993) |
| [43] | Sirovich, L., Turbulence and the dynamics of coherent structures. I. Coherent structures, Quart Appl Math, 45, 3, 561-571 (1987) · Zbl 0676.76047 |
| [44] | Bui-Thanh, T.; Damodaran, M.; Willcox, K., Proper orthogonal decomposition extensions for parametric applications in compressible aerodynamics, (21st AIAA applied aerodynamics conference (2003)) |
| [45] | Dolci, V.; Arina, R., Proper orthogonal decomposition as surrogate model for aerodynamic optimization, Int J Aerospace Eng, 2016, 1-15 (2016) |
| [46] | Dupuis, R.; Jouhaud, J.-C.; Sagaut, P., Aerodynamic data predictions for transonic flows via a machine-learning-based surrogate model, (2018 AIAA/ASCE/AHS/ASC structures, structural dynamics, and materials conference (2018)) |
| [47] | Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O., Scikit-learn: Machine learning in python, J Mach Learn Res, 12, null, 2825-2830 (2011) · Zbl 1280.68189 |
| [48] | Loeppky, J. L.; Sacks, J.; Welch, W. J., Choosing the sample size of a computer experiment: A practical guide, Technometrics, 51, 4, 366-376 (2009) |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
