Abstract
This study focuses on improving the performance of a deep learning neural network to infer Reynolds-averaged Navier-Stokes (RANS) solution, proposed by Thuerey et al. 2019, in the cases of airfoils with high wake formation behind them. The model is based on a U-Net architecture, which calculates pressure and velocity solutions for fluid flow around an airfoil. In this work, we propose on further training an already trained model on a selectively generated data which would be representative of the underperforming test samples. The property that we chose for selectively generating data was the fraction of negative x-velocity in the domain. We were able to observe that using our methods, the performance on the samples with high wake formation (i.e. flow over airfoils at high angle of attack) was improved. But the improved performance came at the cost of performance on samples with lower wake formation. The trend of improvement in samples with high wake formation shows that there is a potential for the model to learn those particular cases.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
A. Voulodimos, N. Doulamis, A. Doulamis, and E. Protopapadakis, “Deep Learning for Computer Vision: A Brief Review,” Comput. Intell. Neurosci., vol. 2018, 2018, https://doi.org/10.1155/2018/7068349.
K. M. Black, H. Law, A. Aldoukhi, J. Deng, and K. R. Ghani, “Deep learning computer vision algorithm for detecting kidney stone composition,” BJU Int., vol. 125, no. 6, pp. 920–924, Jun. 2020, https://doi.org/10.1111/bju.15035.
D. W. Otter, J. R. Medina, and J. K. Kalita, “A Survey of the Usages of Deep Learning for Natural Language Processing,” IEEE Trans. Neural Networks Learn. Syst., vol. 32, no. 2, pp. 604–624, Feb. 2021, https://doi.org/10.1109/TNNLS.2020.2979670.
X. Guo, W. Li, and F. Iorio, “Convolutional neural networks for steady flow approximation,” in Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, vol. 13-17-Augu, pp. 481–490, https://doi.org/10.1145/2939672.2939738.
W. Dong, J. Liu, Z. Xie, and D. Li, “Adaptive neural network-based approximation to accelerate eulerian fluid simulation,” in International Conference for High Performance Computing, Networking, Storage and Analysis, SC, Nov. 2019, pp. 1–22, https://doi.org/10.1145/3295500.3356147.
J. Tompson, K. Schlachter, P. Sprechmann, and K. Perlin, “Accelerating eulerian fluid simulation with convolutional networks,” 34th Int. Conf. Mach. Learn. ICML 2017, vol. 7, pp. 5258–5267, Jul. 2017, Accessed: Aug. 28, 2021. [Online]. Available: https://arxiv.org/abs/1607.03597v6.
R. M. Greenman and K. R. Roth, “High-lift optimization design using neural networks on a multielements airfoil,” J. Fluids Eng. Trans. ASME, vol. 121, no. 2, pp. 434–440, Jun. 1999, https://doi.org/10.1115/1.2822228.
D. Kochkov, J. A. Smith, A. Alieva, Q. Wang, M. P. Brenner, and S. Hoyer, “Machine learning–accelerated computational fluid dynamics,” Proc. Natl. Acad. Sci. U. S. A., vol. 118, no. 21, May 2021, https://doi.org/10.1073/pnas.2101784118.
C. Soize et al., “Enhancing model predictability for a scramjet using probabilistic learning on manifolds,” AIAA J., vol. 57, no. 1, pp. 365–378, Dec. 2019, https://doi.org/10.2514/1.J057069.
V. Sekar, M. Zhang, C. Shu, and B. C. Khoo, “Inverse design of airfoil using a deep convolutional neural network,” AIAA J., vol. 57, no. 3, pp. 993–1003, Jan. 2019, https://doi.org/10.2514/1.J057894.
J. X. Wang, J. L. Wu, and H. Xiao, “Physics-informed machine learning approach for reconstructing Reynolds stress modeling discrepancies based on DNS data,” Phys. Rev. Fluids, vol. 2, no. 3, p. 034603, Mar. 2017, https://doi.org/10.1103/PhysRevFluids.2.034603.
J. X. Wang, J. Huang, L. Duan, and H. Xiao, “Prediction of Reynolds stresses in high-Machnumber turbulent boundary layers using physicsinformed machine learning,” Theor. Comput. Fluid Dyn., vol. 33, no. 1, pp. 1–19, Dec. 2019, https://doi.org/10.1007/s00162-018-0480-2.
N. Thuerey, K. Weißenow, L. Prantl, and X. Hu, “Deep learning methods for reynolds-averaged navier–stokes simulations of airfoil flows,” AIAA J., vol. 58, no. 1, pp. 25–36, Jan. 2020, https://doi.org/10.2514/1.J058291.
O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2015, vol. 9351, pp. 234–241, https://doi.org/10.1007/978-3-319-24574-4_28.
M. S. Selig., “UIUC Airfoil Data Site,” 2012. http://www.ae.illinois.edu/mselig/ads/coord_database.html#N (accessed May 01, 2021).
C. Geuzaine and J. F. Remacle, “Gmsh: A 3-D finite element mesh generator with built-in pre- and postprocessing facilities,” International Journal for Numerical Methods in Engineering, 2009. http://gmsh.info/ (accessed May 01, 2021).
“OpenFOAM | Free CFD Software | The OpenFOAM Foundation,” 2019. https://openfoam.org/ (accessed May 01, 2021).
P. R. Spalart and S. R. Allmaras, “One-equation turbulence model for aerodynamic flows,” in Recherche aerospatiale, Jan. 1994, no. 1, pp. 5–21, https://doi.org/10.2514/6.1992-439.
A. Paszke et al., “PyTorch: An imperative style, highperformance deep learning library,” Adv. Neural Inf. Process. Syst., vol. 32, Dec. 2019, [Online]. Available: http://arxiv.org/abs/1912.01703.
D. P. Kingma and J. L. Ba, “Adam: A method for stochastic optimization,” 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc., Dec. 2015, [Online]. Available: http://arxiv.org/abs/1412.6980.
Acknowledgements
We would like to thank Thuerey et al. for making the code used in their work [13] available on https://github.com/thunil/Deep-Flow-Prediction. The second author A.A. would like to thank the Department of Science and Technology, Government of India, for the financial support under Grant No. DST/INSPIRE/04/2019/001001.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Ananthajit, A., Assam, A. (2023). Training of Neural Network on Selectively Generated Data for Flow over Airfoils at Higher Angle of Attack. In: Bhattacharyya, S., Verma, S., Harikrishnan, A.R. (eds) Fluid Mechanics and Fluid Power (Vol. 3). FMFP 2021. Lecture Notes in Mechanical Engineering. Springer, Singapore. https://doi.org/10.1007/978-981-19-6270-7_1
Download citation
DOI: https://doi.org/10.1007/978-981-19-6270-7_1
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-6269-1
Online ISBN: 978-981-19-6270-7
eBook Packages: EngineeringEngineering (R0)