Abstract
Physics-Informed Neural Networks (PINNs) are machine learning tools that approximate the solution of general partial differential equations (PDEs) by adding them in some form as terms of the loss/cost function of a Neural Network. Most pieces of work in the area of PINNs tackle non-linear PDEs. Nevertheless, many interesting problems involving linear PDEs may benefit from PINNs; these include parametric studies, multi-query problems, and parabolic (transient) PDEs. The purpose of this paper is to explore PINNs for linear PDEs whose solutions may present one or more boundary layers. More specifically, we analyze the steady-state reaction-advection-diffusion equation in regimes in which the diffusive coefficient is small in comparison with the reactive or advective coefficients. We show that adding information about these coefficients as predictor variables in a PINN results in better prediction models than a PINN that only uses spatial information as predictor variables. Even though using these coefficients when training a PINN model is a common strategy for inverse problems, to the best of our knowledge we are the first to consider these coefficients for parametric direct problems. This finding may be instrumental in multiscale problems where the coefficients of the PDEs present high variability in small spatiotemporal regions of the domain, and therefore PINNs may be employed together with domain decomposition techniques to efficiently approximate the PDEs locally at each partition of the spatiotemporal domain, without resorting to different learned PINN models at each of these partitions.
The authors are presented in alphabetical order.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The experiments presented in this paper can be reproduced in Google Colaboratory:
https://colab.research.google.com/drive/1dzzK41xIrmi5ozzO4IzBnkktGjI90j_-.
References
Barrenechea, G., Jaillet, F., Paredes, D., Valentin, F.: The multiscale hybrid mixed method in general polygonal meshes. Numerische Mathematik 145(1), 197–237 (2020). https://doi.org/10.1007/s00211-020-01103-5
Berg, J., Nyström, K.: A unified deep artificial neural network approach to partial differential equations in complex geometries. Neurocomputing 317, 28–41 (2018)
Cai, S., Wang, Z., Wang, S., Perdikaris, P., Karniadakis, G.E.: Physics-informed neural networks for heat transfer problems. J. Heat Transfer 143(6), 060801 (2021)
Capuano, G., Rimoli, J.J.: Smart finite elements: a novel machine learning application. Comput. Methods Appl. Mech. Eng. 345, 363–381 (2019)
Chan, S., Elsheikh, A.H.: A machine learning approach for efficient uncertainty quantification using multiscale methods. J. Comput. Phys. 354, 493–511 (2018)
E, W., Yu, B.: The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems. Commun. Math. Stat. 6(1), 1–12 (2018). https://doi.org/10.1007/s40304-018-0127-z
Fabian, J.H.L., Gomes, A.T.A., Ogasawara, E.: Estimating the execution time of fully-online multiscale numerical simulations. In: Proceedings of the XXI Brazilian Symposium on High-Performance Computing Systems (WSCAD), pp. 191–202. Sociedade Brasileira de Computação (2020)
Fabian, J.H.L., Gomes, A.T.A., Ogasawara, E.: Estimating the execution time of the coupled stage in multiscale numerical simulations. In: Nesmachnow, S., Castro, H., Tchernykh, A. (eds.) CARLA 2020. CCIS, vol. 1327, pp. 86–100. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-68035-0_7
Harder, C., Paredes, D., Valentin, F.: On a multiscale hybrid-mixed method for advective-reactive dominated problems with heterogeneous coefficients. Multiscale Model. Simul. 13(2), 491–518 (2015)
Jagtap, A., Karniadakis, G.: Extended physics-informed neural networks (XPINNs): a generalized space-time domain decomposition based deep learning framework for nonlinear partial differential equations. Commun. Comput. Phys. 28(5), 2002–2041 (2020). https://global-sci.org/intro/article_detail/cicp/18403.html
Jagtap, A.D., Kharazmi, E., Karniadakis, G.E.: Conservative physics-informed neural networks on discrete domains for conservation laws: applications to forward and inverse problems. Comput. Methods Appl. Mech. Eng. 365, 113028 (2020). https://www.sciencedirect.com/science/article/pii/S0045782520302127
Jagtap, A.D., Karniadakis, G.E.: Extended physics-informed neural networks (XPINNs): a generalized space-time domain decomposition based deep learning framework for nonlinear partial differential equations. In: AAAI Spring Symposium: MLPS (2021)
Karpatne, A., Watkins, W., Read, J., Kumar, V.: Physics-guided neural networks (PGNN): an application in lake temperature modeling. arXiv preprint arXiv:1710.11431 (2017)
Kharazmi, E., Zhang, Z., Karniadakis, G.E.: Variational physics-informed neural networks for solving partial differential equations (2019). https://arxiv.org/abs/1912.00873
Lagaris, I.E., Likas, A., Fotiadis, D.I.: Artificial neural networks for solving ordinary and partial differential equations. IEEE Trans. Neural Netw. 9(5), 987–1000 (1998)
Mao, Z., Jagtap, A.D., Karniadakis, G.E.: Physics-informed neural networks for high-speed flows. Comput. Methods Appl. Mech. Eng. 360, 112789 (2020)
Liao, Y., Ming, P.: Deep Nitsche method: deep Ritz method with essential boundary conditions. Commun. Comput. Phys. 29(5), 1365–1384 (2021). https://doi.org/10.4208/cicp.OA-2020-0219
Misyris, G.S., Venzke, A., Chatzivasileiadis, S.: Physics-informed neural networks for power systems. In: 2020 IEEE Power & Energy Society General Meeting (PESGM), pp. 1–5. IEEE (2020)
Raissi, M., Perdikaris, P., Karniadakis, G.: 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). https://www.sciencedirect.com/science/article/pii/S0021999118307125
Raissi, M., Karniadakis, G.E.: Hidden physics models: machine learning of nonlinear partial differential equations. J. Comput. Phys. 357, 125–141 (2018)
Sirignano, J., Spiliopoulos, K.: DGM: a deep learning algorithm for solving partial differential equations. J. Comput. Phys. 375, 1339–1364 (2018)
Yeung, T.S.A., Chung, E.T., See, S.: A deep learning based nonlinear upscaling method for transport equations. arXiv preprint arXiv:2007.03432 (2020)
Acknowledgment
The authors were supported by CAPES/Brazil under Project EOLIS (No. 88881.520197/2020-01). Frédéric Valentin was supported by Inria/France under the Inria International Chair, by CNPq/Brazil under Project No. 309173/2020-5, and by FAPERJ/Brazil under Project No. E-26/201.182/2021.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Gomes, A.T.A., da Silva, L.M., Valentin, F. (2022). Improving Boundary Layer Predictions Using Parametric Physics-Aware Neural Networks. In: Navaux, P., Barrios H., C.J., Osthoff, C., Guerrero, G. (eds) High Performance Computing. CARLA 2022. Communications in Computer and Information Science, vol 1660. Springer, Cham. https://doi.org/10.1007/978-3-031-23821-5_7
Download citation
DOI: https://doi.org/10.1007/978-3-031-23821-5_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-23820-8
Online ISBN: 978-3-031-23821-5
eBook Packages: Computer ScienceComputer Science (R0)