Abstract
Molecular-continuum coupled flow simulations are used in many applications to build a bridge across spatial or temporal scales. Hence, they allow to investigate effects beyond flow scenarios modeled by any single-scale method alone, such as a discrete particle system or a partial differential equation solver. On the particle side of the coupling, often molecular dynamics (MD) is used to obtain trajectories based on pairwise molecule interaction potentials. However, since MD is computationally expensive and macroscopic flow quantities sampled from MD systems often highly fluctuate due to thermal noise, the applicability of molecular-continuum methods is limited. If machine learning (ML) methods can learn and predict MD based flow data, then this can be used as a noise filter or even to replace MD computations, both of which can generate tremendous speed-up of molecular-continuum simulations, enabling emerging applications on the horizon.
In this paper, we develop an advanced hybrid ML model for MD data in the context of coupled molecular-continuum flow simulations: A convolutional autoencoder deals with the spatial extent of the flow data, while a recurrent neural network is used to capture its temporal correlation. We use the open source coupling tool MaMiCo to generate MD datasets for ML training and implement the hybrid model as a PyTorch-based filtering module for MaMiCo. It is trained with real MD data from different flow scenarios including a Couette flow validation setup and a three-dimensional vortex street. Our results show that the hybrid model is able to learn and predict smooth flow quantities, even for very noisy MD input data. We furthermore demonstrate that also the more complex vortex street flow data can accurately be reproduced by the ML module.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bauer, M., Köstler, H., Rüde, U.: lbmpy: automatic code generation for efficient parallel lattice Boltzmann methods. J. Comput. Sci. 49, 101269 (2021)
Borg, M.K., Lockerby, D.A., Ritos, K., Reese, J.M.: Multiscale simulation of water flow through laboratory-scale nanotube membranes. J. Membrane Sci. 567, 115–126 (2018). ISSN 0376-7388
Bungartz, H.J., et al.: preCICE - a fully parallel library for multi-physics surface coupling. Comput. I & Fluids 141, 250–258 (2016)
Fukami, K., Hasegawa, K., Nakamura, T., Morimoto, M., Fukagata, K.: Model order reduction with neural networks: application to laminar and turbulent flows. SN Comput. Sci. 2, 1–16 (2021)
Grinberg, L.: Proper orthogonal decomposition of atomistic flow simulations. J. Comput. Phys. 231(16), 5542–5556 (2012)
Jarmatz, P., Maurer, F., Wittenberg, H., Neumann, P.: MaMiCo: non-local means and pod filtering with flexible data-flow for two-way coupled molecular-continuum HPC flow simulation. J. Comput. Sci. 61, 101617 (2022)
Jarmatz, P., Neumann, P.: MaMiCo: parallel noise reduction for multi-instance molecular-continuum flow simulation. In: Rodrigues, J.M.F., et al. (eds.) ICCS 2019. LNCS, vol. 11539, pp. 451–464. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-22747-0_34
Jarmatz, P., et al.: MaMiCo 2.0: an enhanced open-source framework for high-performance molecular-continuum flow simulation. SoftwareX 20, 101251 (2022). ISSN 2352-7110
Kadupitiya, J., Sun, F., Fox, G., Jadhao, V.: Machine learning surrogates for molecular dynamics simulations of soft materials. J. Comput. Sci. 42, 101107 (2020)
Lecun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998). https://doi.org/10.1109/5.726791
Nakamura, T., Fukagata, K.: Robust training approach of neural networks for fluid flow state estimations. Int. J. Heat Fluid Flow 96, 108997 (2022)
Nakamura, T., Fukami, K., Hasegawa, K., Nabae, Y., Fukagata, K.: Convolutional neural network and long short-term memory based reduced order surrogate for minimal turbulent channel flow. Phys. Fluids 33(2), 025116 (2021). https://doi.org/10.1063/5.0039845
Neumann, P., Bian, X.: MaMiCo: transient multi-instance molecular-continuum flow simulation on supercomputers. Comput. Phys. Commun. 220, 390–402 (2017)
Niethammer, C., et al.: ls1 mardyn: the massively parallel molecular dynamics code for large systems. J. Chem. Theory Comput. 10(10), 4455–4464 (2014)
Ren, X.G., Wang, Q., Xu, L.Y., Yang, W.J., Xu, X.H.: HACPar: an efficient parallel multiscale framework for hybrid atomistic-continuum simulation at the micro-and nanoscale. Adv. Mech. Eng. 9(8) (2017)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Schäfer, M., Turek, S., Durst, F., Krause, E., Rannacher, R.: Benchmark computations of laminar flow around a cylinder. In: Hirschel, E.H. (ed.) Flow Simulation with High-Performance Computers II. NNFM, pp. 547–566. Springer, Cham (1996). https://doi.org/10.1007/978-3-322-89849-4_39
Smith, E.: On the coupling of molecular dynamics to continuum computational fluid dynamics. Sch. Mech. Eng. (2013)
Tang, Y.H., Kudo, S., Bian, X., Li, Z., Karniadakis, G.E.: Multiscale universal interface: a concurrent framework for coupling heterogeneous solvers. J. Comput. Phys. 297, 13–31 (2015)
Thomas, M., Corry, B.: A computational assessment of the permeability and salt rejection of carbon nanotube membranes and their application to water desalination. Philos. Trans. R. Soc. A: Math. Phys. Eng. Sci. 374(2060), 20150020 (2016)
Veen, L.E., Hoekstra, A.G.: Easing multiscale model design and coupling with MUSCLE 3. In: Krzhizhanovskaya, V.V., et al. (eds.) ICCS 2020. LNCS, vol. 12142, pp. 425–438. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-50433-5_33
Wiewel, S., Becher, M., Thuerey, N.: Latent space physics: towards learning the temporal evolution of fluid flow. In: Computer Graphics Forum, vol. 38, pp. 71–82. Wiley Online Library (2019)
Wittenberg, H., Neumann, P.: Transient two-way molecular-continuum coupling with OpenFOAM and MaMiCo: a sensitivity study. Computation 9(12) (2021). https://doi.org/10.3390/computation9120128. ISSN 2079-3197
Yin, W., Kann, K., Yu, M., Schütze, H.: Comparative study of CNN and RNN for natural language processing. arXiv:1702.01923 (2017)
Zhang, A., Lipton, Z.C., Li, M., Smola, A.J.: Dive into deep learning. arXiv preprint arXiv:2106.11342 (2021)
Acknowledgments
We thank the projects “hpc.bw” and “MaST” of dtec.bw – Digitalization and Technology Research Center of the Bundeswehr for providing computational resources, as the HPC cluster “HSUper” has been used to train and validate the ML models presented in this paper. We also want to thank Prof. Zhen Li and the MuthComp Group of Clemson University for fruitful discussions and exchange of ideas as well as provision of office space and IT systems.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Jarmatz, P., Lerdo, S., Neumann, P. (2023). Convolutional Recurrent Autoencoder for Molecular-Continuum Coupling. In: Mikyška, J., de Mulatier, C., Paszynski, M., Krzhizhanovskaya, V.V., Dongarra, J.J., Sloot, P.M. (eds) Computational Science – ICCS 2023. ICCS 2023. Lecture Notes in Computer Science, vol 10476. Springer, Cham. https://doi.org/10.1007/978-3-031-36027-5_42
Download citation
DOI: https://doi.org/10.1007/978-3-031-36027-5_42
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-36026-8
Online ISBN: 978-3-031-36027-5
eBook Packages: Computer ScienceComputer Science (R0)