Communications on Applied Mathematics and Computation >

2024 , Vol. 6 >Issue 3: 1832 - 1859

DOI: https://doi.org/10.1007/s42967-023-00334-1

ORIGINAL PAPERS

Machine Learning Approaches for the Solution of the Riemann Problem in Fluid Dynamics: a Case Study

Expand
  • 1 Los Alamos National Laboratory, Applied Mathematics and Plasma Physics (T-5), Los Alamos, USA;
    2 Los Alamos National Laboratory, Continuum Models and Numerical Methods (XCP-4), Los Alamos, USA;
    3 Los Alamos National Laboratory, Space Data Science and Systems (ISR-3), Los Alamos, USA

Received date: 2022-12-26

  Revised date: 2022-12-26

  Accepted date: 2023-09-11

  Online published: 2024-12-20

Supported by

This work was performed under the auspices of the National Nuclear Security Administration of the US Department of Energy at Los Alamos National Laboratory under Contract No. DE-AC52-06NA25396. The authors gratefully acknowledge the support of the US Department of Energy National Nuclear Security Administration Advanced Simulation and Computing Program. The Los Alamos unlimited release number is LA-UR-19-32257. The authors would like to thank J. Albright for numerous discussions and his valuable comments.

Fold

Abstract

We present our results by using a machine learning (ML) approach for the solution of the Riemann problem for the Euler equations of fluid dynamics. The Riemann problem is an initial-value problem with piecewise-constant initial data and it represents a mathematical model of the shock tube. The solution of the Riemann problem is the building block for many numerical algorithms in computational fluid dynamics, such as finite-volume or discontinuous Galerkin methods. Therefore, a fast and accurate approximation of the solution of the Riemann problem and construction of the associated numerical fluxes is of crucial importance. The exact solution of the shock tube problem is fully described by the intermediate pressure and mathematically reduces to finding a solution of a nonlinear equation. Prior to delving into the complexities of ML for the Riemann problem, we consider a much simpler formulation, yet very informative, problem of learning roots of quadratic equations based on their coefficients. We compare two approaches: (i) Gaussian process (GP) regressions, and (ii) neural network (NN) approximations. Among these approaches, NNs prove to be more robust and efficient, although GP can be appreciably more accurate (about 30%). We then use our experience with the quadratic equation to apply the GP and NN approaches to learn the exact solution of the Riemann problem from the initial data or coefficients of the gas equation of state (EOS). We compare GP and NN approximations in both regression and classification analysis and discuss the potential benefits and drawbacks of the ML approach.

Key words： Machine learning (ML); Neural network (NN); Gaussian process (GP); Riemann problem; Numerical fluxes; Finite-volume method

Cite this article

Vitaly Gyrya, Mikhail Shashkov, Alexei Skurikhin, Svetlana Tokareva . Machine Learning Approaches for the Solution of the Riemann Problem in Fluid Dynamics: a Case Study[J]. Communications on Applied Mathematics and Computation, 2024 , 6(3) : 1832 -1859 . DOI: 10.1007/s42967-023-00334-1

References

1. Machine Learning for Computational Fluid and Solid Dynamics. Santa Fe, NM, USA, 19–21 (2019)
2. Scientific Machine Learning. ICERM, Providence, RI, USA, January 28–30 (2019)
3. Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G.S., Davis, A., Dean, J., Devin, M., Ghemawat, S., Goodfellow, I., Harp, A., Irving, G., Isard, M., Jia, Y., Jozefowicz, R., Kaiser, L., Kudlur, M., Levenberg, J., Mané, D., Monga, R., Moore, S., Murray, D., Olah, C., Schuster, M., Shlens, J., Steiner, B., Sutskever, I., Talwar, K., Tucker, P., Vanhoucke, V., Vasudevan, V., Viégas, F., Vinyals, O., Warden, P., Wattenberg, M., Wicke, M., Yu, Y., Zheng, X.: TensorFlow: large-scale machine learning on heterogeneous systems (2015). Software available from tensorflow.org
4. Bar-Sinai, Y., Hoyer, S., Hickey, J., Brenner, M.P.: Learning data-driven discretizations for partial differential equations. PNAS 116, 15344–15349 (2019)
5. Beck, C., E, W.N., Jentzen, A.: Machine learning approximation algorithms for high-dimensional fully nonlinear partial differential equations and second-order backward stochastic differential equations. Journal of Nonlinear Science 29, 1563–1619 (2019)
6. Berg, J., Nystróm, K.: A unified deep artificial neural network approach to partial differential equations in complex geometries. Neurocomputing 317, 28–41 (2018)
7. Brunton, S., Proctor, J., Kutz, N.: Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proc. Natl. Acad. Sci. 113(15), 3932–3937 (2016)
8. Cotter, N.E.: The Stone-Weierstrass theorem and its application to neural networks. IEEE Trans. Neural Networks 1, 290–295 (1990)
9. Csái, B.C.: Approximation with artificial neural networks. MSc Thesis, Eotvos Lorand University (ELTE), Budapest, Hungary (2001)
10. Du, J., Xu, Y.: Hierarchical deep neural network for multivariate regression. Pattern Recogn. 63, 149–157 (2017)
11. François, F., et al.: Keras (2015) https:// keras. io
12. Golak, S.: A MLP solver for first and second order partial differential equations. In: de Sá, J.M., Alexandre, L.A., Duch, W., Mandic, D. (eds.) Artificial Neural Networks-ICANN 2007, pp. 789–797. Springer-Verlag, Berlin, Heidelberg (2007)
13. Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press, Cambridge, MA (2016)
14. Goulianas, K., Margaris, A., Refanidis, I., Diamantaras, K., Papadimitriou, T.: A back propagationtype neural network architecture for solving the complete n×n nonlinear algebraic system of equations. Adv. Pure Math. 6, 455–480 (2016)
15. Gyrya, V., Shashkov, M., Skurikhin, A.: Exploration of machine learning for polynomial root finding. https:// www. resea rchga te. netpu blica tion/ 33110 1795_ Explo ration_ of_ Machi ne_ learn ing_ for_ Polyn omial_ Root_ Findi ng_ Motiv ation
16. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition, pp. 770–778. IEEE (2016)
17. Hornik, K., Stinchcombe, M., White, H.: Multilayer feedforward networks are universal approximators. Neural Netw. 2, 359–366 (1989)
18. Huang, D.S., Ip, H.H.S., Chi, Z.: A neural root finder of polynomials based on root moments. Neural Comput. 16, 1721–1762 (2004)
19. Kumar, M., Yadav, N.: Multilayer perceptrons and radial basis function neural network methods for the solution of differential equations: a survey. Comput. Math. Appl. 62, 3796–3811 (2011)
20. Kurková, V.: Kolmogorov’s theorem and multilayer neural networks. Neural Netw. 5, 501–506 (1992)
21. Lagaris, I.E., Likas, A., Fotiadis, D.I.: Artificial neural networks for solving ordinary and partial differential equations. IEEE Trans. Neural Networks 9, 987–1000 (1998)
22. LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521, 436–444 (2015)
23. Lin, H., Jegelka, S.: ResNet with one-neuron hidden layers is a universal approximator. In: Proc. of the Intern. Conf. on Neural Information Processing Systems, pp. 6172–6181 (2018)
24. Long, Z., Lu, Y., Ma, X., Dong, B.: PDE-Net: learning PDEs from data. In: Proc. the 35th Intern. Conf. on Machine Learning. PMLR (2018)
25. Lu, Z., Pu, H., Wang, F., Hu, Z., Wang, L.: The expressive power of neural networks: a view from width. In: Proc. of the Intern. Conf. on Neural Information Processing Systems, pp. 6232–6240 (2017)
26. Lye, K. O., Mishra, S., Ray, D.: Deep learning observables in computational fluid dynamics. J. Comput. Phys. 410, 109339 (2020)
27. Meade, A.J., Fernandez, A.A.: Solution of nonlinear ordinary differential equations by feedforward neural networks. Mathl. Comput. Modeling 20, 19–44 (1994)
28. Mishra, S.: A machine learning framework for data driven acceleration of computations of differential equations. Mathematics in Engineering 1, 118–146 (2018)
29. Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., Duchesnay, E.: Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011)
30. Qin, T., Wu, K., Xiu, D.: Data driven governing equations approximation using deep neural networks. J. Comput. Phys. 395, 620–635 (2019)
31. Raissi, M., Babee, H., Karniadakis, G. E.: Parametric Gaussian process regression for big data. Comput. Mech. 64, 409–416 (2019)
32. Raissi, M., Karniadakis, G.E.: Hidden physics models: machine learning of nonlinear partial differential equations. J. Comput. Phys. 357, 125–141 (2018)
33. Raissi, M., Perdikaris, P., Karniadakis, G.E.: Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J. Comput. Phys. 378, 686–707 (2019)
34. Rasmussen, C.E., Williams, C.K.I.: Gaussian Processes for Machine Learning. Adaptive Computation and Machine Learning. MIT Press, Cambridge, MA (2006)
35. Ray, D., Hesthaven, J.: An artificial neural network as a troubled-cell indicator. J. Comput. Phys. 367, 166–191 (2018)
36. Ray, D., Hesthaven, J.: Detecting troubled-cells on two-dimensional unstructured grids using a neural network. J. Comput. Phys. 397, 108845 (2018)
37. Rumelhart, D. E., McClelland, J.L.: PDP Research Group. Parallel Distributed Processing: Explorations in the Microstructure of Cognition. Vol. 1: Foundations. MIT Press, Cambridge, MA (1986)
38. Sirignano, J., Spiliopoulos, K.: DGM: a deep learning algorithm for solving partial differential equations. J. Comput. Phys. 375, 1339–1364 (2018)
39. Specht, D.F.: A general regression neural network. IEEE Trans. Neural Networks 2, 568–576 (1991)
40. Tokareva, S., Shashkov, M., Skurikhin, A.: Machine learning approach for the solution of the Riemann problem in fluid dynamics. https:// www. resea rchga te. net/ publi cation/ 33079 8897_ Machi ne_ learn ing_appro ach_ for_ the_ solut ion_ of_ the_ Riema nn_ probl em_ in_ fluid_ dynam ics
41. Tompson, J., Schlachter, K., Sprechmann, P., Perlin, K.: Accelerating Eulerian fluid simulation with convolutional networks. In: Proc. 34th Intern. Conf. on Machine Learning (2017)
42. Toro, E. F.: Riemann Solvers and Numerical Methods for Fluid Dynamics. Springer-Verlag, Berlin, Heidelberg (2009)
43. Veiga, H.M., Abgrall, R.: Towards a general stabilisation method for conservation laws using a multilayer perceptron neural network: 1D scalar and system of equations. In: ECCM-ECFD 2018 6th European Conference on Computational Mechanics (Solids, Structures and Coupled Problems) 7th European Conference on Computational Fluid Dynamics, Glasgow, United Kingdom (2018)
44. Winovich, N., Ramani, K., Lin, G.: ConvPDE-UQ: convolutional neural networks with quantified uncertainty for heterogeneous elliptic partial differential equations on varied domains. 394, 263–279 (2019)
45. Xie, Y., Franz, E., Chu, M., Thuerey, N.: tempoGAN: a temporally coherent, volumetric GAN for super-resolution fluid flow. ACM Transactions on Graphics 37(95), 1–15 (2018)
Options
Outlines

/

〈 〉

Copyright © Editorial By Communications on Applied Mathematics and Computation
沪ICP备09014157