Towards real-time fluid dynamics simulation: a data-driven NN-MPS method and its implementation. (English) Zbl 1530.76054
Summary: In this work, we construct a data-driven model to address the computing performance problem of the moving particle semi-implicit method by combining the physics intuition of the method with a machine-learning algorithm. A fully connected artificial neural network is implemented to solve the pressure Poisson equation, which is reformulated as a regression problem. We design context-based feature vectors for particle-based on the Poisson equation. The neural network maintains the original particle method’s accuracy and stability, while drastically accelerates the pressure calculation. It is very suitable for GPU parallelization, edge computing scenarios and real-time simulations.
MSC:
| 76M28 | Particle methods and lattice-gas methods |
| 76M99 | Basic methods in fluid mechanics |
| 76D05 | Navier-Stokes equations for incompressible viscous fluids |
| 68T05 | Learning and adaptive systems in artificial intelligence |
| 68T07 | Artificial neural networks and deep learning |
Keywords:
Lagrangian moving particle semi-implicit method; machine-learning algorithm; fully connected artificial neural network; edge computing; Navier-Stokes equations; pressure Poisson equation; dam collapse problemReferences:
| [1] | Koshizuka, S.; Oka, Y., Moving-particlesemi-implicit method for fragmentation of incompressible fluid, Nucl. Sci. Eng, 123, 3, 421-434 (1996) · doi:10.13182/NSE96-A24205 |
| [2] | Monaghan, J. J., Smoothed particle hydrodynamics, Annu. Rev. Astron. Astrophys., 30, 1, 543-574 (1992) · doi:10.1146/annurev.aa.30.090192.002551 |
| [3] | Gambaruto, A. M., Computational haemodynamics of small vessels using the moving particle semi-implicit (mps) method, J. Comput. Phys., 302, 68-96 (2015) · Zbl 1349.76937 · doi:10.1016/j.jcp.2015.08.039 |
| [4] | Murotani, K.; Koshizuka, S.; Tamai, T.; Shibata, K.; Mitsume, N.; Yoshimura, S.; Tanaka, S.; Hasegawa, K.; Nagai, E.; Fujisawa, T., Development of hierarchical domain decomposition explicit mps method and application to large-scale tsunami analysis with floating objects, Adv. Model. Simul. Eng. Sci., 1, 1, 16-35 (2014) · doi:10.15748/jasse.1.16 |
| [5] | Okabe, T.; Matsutani, H.; Honda, T.; Yashiro, S., Numerical simulation of microscopic flow in a fiber bundle using the moving particle semi-implicit method, composites part A, Appl. Sci. Manuf., 43, 10, 1765-1774 (2012) · doi:10.1016/j.compositesa.2012.05.003 |
| [6] | Duan, G.; Chen, B., Comparison of parallel solvers for moving particle semi-implicit method, Eng. Comput., 32, 3, 834-862 (2015) · doi:10.1108/EC-02-2014-0029 |
| [7] | Koshizuka, S.; Shibata, K.; Kondo, M.; Matsunaga, T., Moving Particle Semi-Implicit Method: A Meshfree Particle Method for Fluid Dynamics (2018), Cambridge, Massachusetts: Academic Press, Cambridge, Massachusetts |
| [8] | Noack, B. R.; Eckelmann, H., A low-dimensional galerkin method for the three-dimensional flow around a circular cylinder, Phys. Fluids, 6, 1, 124-143 (1994) · Zbl 0826.76071 · doi:10.1063/1.868433 |
| [9] | Treuille, A.; Lewis, A.; Popović, Z., Model reduction for real-time fluids, ACM Trans. Graph., 25, 3, 826-834 (2006) · doi:10.1145/1141911.1141962 |
| [10] | Ando, R.; Thürey, N.; Wojtan, C., Computer Graphics Forum, 34, A dimension-reduced pressure solver for liquid simulations, 473-480 (2015), Hoboken, New Jersey, U.S: Wiley Online Library, Hoboken, New Jersey, U.S |
| [11] | De Witt, T.; Lessig, C.; Fiume, E., Fluid simulation using Laplacian eigenfunctions, ACM Trans. Graph., 31, 1, 1-11 (2012) · doi:10.1145/2077341.2077351 |
| [12] | Li, S.; Duan, G.; Sakai, M., Development of a reduced-order model for large-scale Eulerian-Lagrangian simulations, Adv. Powder Technol., 33, 8, 103632 (2022) · doi:10.1016/j.apt.2022.103632 |
| [13] | Ladickỳ, L.; Jeong, S.; Solenthaler, B.; Pollefeys, M.; Gross, M., Data-driven fluid simulations using regression forests, ACM Trans. Graph., 34, 6, 1-9 (2015) · doi:10.1145/2816795.2818129 |
| [14] | Li, Z.; Farimani, A. B., Graph neural network-accelerated Lagrangian fluid simulation, Comput. Graph, 103, 201-211 (2022) · doi:10.1016/j.cag.2022.02.004 |
| [15] | Bai, J.; Zhou, Y.; Ma, Y.; Jeong, H.; Zhan, H.; Rathnayaka, C.; Sauret, E.; Gu, Y., A general neural particle method for hydrodynamics modeling, Comput. Methods Appl. Mech. Eng., 393, 114740 (2022) · Zbl 1507.76153 · doi:10.1016/j.cma.2022.114740 |
| [16] | Yang, C.; Yang, X.; Xiao, X., Data-driven projection method in fluid simulation, Comput. Animat Virtual. Worlds, 27, 3-4, 415-424 (2016) · doi:10.1002/cav.1695 |
| [17] | Tompson, J.; Schlachter, K.; Sprechmann, P.; Perlin, K., Accelerating Eulerian fluid simulation with convolutional networks, 3424-3433 (2017) |
| [18] | Martin, J. C.; Moyce, W. J.; Martin, J.; Moyce, W.; Penney, W. G.; Price, A.; Thornhill, C., Part iv. an experimental study of the collapse of liquid columns on a rigid horizontal plane, Philos Trans. R. Soc. Lond A, 244, 312-324 (1952) · Zbl 0046.19807 |
| [19] | Hernquist, L.; Katz, N., Treesph-a unification of sph with the hierarchical tree method, Astrophys. J., Suppl. Ser., 70, 419-446 (1989) · doi:10.1086/191344 |
| [20] | Lee, B.-H.; Park, J.-C.; Kim, M.-H.; Hwang, S.-C., Step-by-step improvement of mps method in simulating violent free-surface motions and impact-loads, Comput. Methods Appl. Mech. Eng., 200, 9-12, 9-12 (2011) · Zbl 1225.76234 · doi:10.1016/j.cma.2010.12.001 |
| [21] | Nair, V.; Hinton, G. E., Rectified linear units improve restricted Boltzmann machines, ICML’10: Proceedings of the 27th International Conference on International Conference on Machine Learning, icml10, 807-814 (2010) |
| [22] | Rumelhart, D. E.; Hinton, G. E.; Williams, R. J., Learning representations by back-propagating errors, nature, 323, 6088, 533-536 (1986) · Zbl 1369.68284 · doi:10.1038/323533a0 |
| [23] | Kingma, D. P.; Ba, J., Adam: A method for stochastic optimization, ICLR (Poster) 2015 (2015), Ithaca, NY: arXiv.org |
| [24] | Srivastava, N.; Hinton, G.; Krizhevsky, A.; Sutskever, I.; Salakhutdinov, R., Dropout: A simple way to prevent neural networks from overfitting, J. Mach. Learn. Res., 15, 1929-1958 (2014) · Zbl 1318.68153 |
| [25] | Abadi, M.; Barham, P.; Chen, J.; Chen, Z.; Davis, A.; Dean, J.; Devin, M.; Ghemawat, S.; Irving, G.; Isard, M.; Manjunath, Kudlur; Josh, Levenberg; Rajat, Monga; Sherry, Moore; Derek, G.; Murray, Benoit; Steiner, Paul; Tucker, Vijay; Vasudevan, Pete; Warden, Martin Wicke; Yuan Yu, Xiaoqiang Zheng, TensorFlow: A system for large-scale machine learning, 16, 265-283 (2016) |
| [26] | Wang, L.; Khayyer, A.; Gotoh, H.; Jiang, Q.; Zhang, C., Enhancement of pressure calculation in projection-based particle methods by incorporation of background mesh scheme, Appl. Ocean Res., 86, 320-339 (2019) · doi:10.1016/j.apor.2019.01.017 |
| [27] | Duan, G.; Koshizuka, S.; Yamaji, A.; Chen, B.; Li, X.; Tamai, T., An accurate and stable multiphase moving particle semi-implicit method based on a corrective matrix for all particle interaction models, Int. J. Numer Methods Eng., 115, 10, 1287-1314 (2018) · Zbl 07865129 · doi:10.1002/nme.5844 |
| [28] | Duan, G.; Matsunaga, T.; Koshizuka, S.; Yamaguchi, A.; Sakai, M., New insights into error accumulation due to biased particle distribution in semi-implicit particle methods, Comput. Methods Appl. Mech. Eng., 388, 114219 (2022) · Zbl 1507.76156 · doi:10.1016/j.cma.2021.114219 |
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.
