Physics-Informed Neural Network Solution of Thermo–Hydro–Mechanical Processes in Porous Media
Publication: Journal of Engineering Mechanics
Volume 148, Issue 11
Abstract
Physics-informed neural networks (PINNs) have received increased interest for forward, inverse, and surrogate modeling of problems described by partial differential equations (PDEs). However, their application to multiphysics problem, governed by several coupled PDEs, presents unique challenges that have hindered the robustness and widespread applicability of this approach. Here we investigate the application of PINNs to the forward solution of problems involving thermo–hydro–mechanical (THM) processes in porous media that exhibit disparate spatial and temporal scales in thermal conductivity, hydraulic permeability, and elasticity. In addition, PINNs are faced with the challenges of the multiobjective and nonconvex nature of the optimization problem. To address these fundamental issues, we (1) rewrote the THM governing equations in dimensionless form that is best suited for deep learning algorithms, (2) propose a sequential training strategy that circumvents the need for a simultaneous solution of the multiphysics problem and facilitates the task of optimizers in the solution search, and (3) leveraged adaptive weight strategies to overcome the stiffness in the gradient flow of the multiobjective optimization problem. Finally, we applied this framework to the solution of several synthetic problems in one and two dimensions.
Get full access to this article
View all available purchase options and get full access to this article.
Data Availability Statement
All data, models, or code generated or used during the study are available in a repository online (https://github.com/sciann/sciann-applications/tree/master/SciANN-PoroElasticity) in accordance with funder data retention policies. The SciANN program can be found online (https://github.com/sciann/sciann-applications).
References
Abadi, M., et al. 2016. “TensorFlow: A system for large-scale machine learning.” In Proc., 12th USENIX Symp. on Operating Systems Design and Implementation (OSDI 16), 265–283. Berkeley, CA: USENIX Association. https://www.usenix.org/conference/osdi16/technical-sessions/presentation/abadi.
Almajid, M. M., and M. O. Abu-Al-Saud. 2021. “Prediction of porous media fluid flow using physics informed neural networks.” J. Pet. Sci. Eng. 208 (Jan): 109205. https://doi.org/10.1016/j.petrol.2021.109205.
Armero, F., and J. Simo. 1992. “A new unconditionally stable fractional step method for non-linear coupled thermomechanical problems.” Int. J. Numer. Methods Eng. 35 (4): 737–766. https://doi.org/10.1002/nme.1620350408.
Ballarini, E., B. Graupner, and S. Bauer. 2017. “Thermal–hydraulic–mechanical behavior of bentonite and sand-bentonite materials as seal for a nuclear waste repository: Numerical simulation of column experiments.” Appl. Clay Sci. 135 (Jan): 289–299. https://doi.org/10.1016/j.clay.2016.10.007.
Baydin, A. G., B. A. Pearlmutter, A. A. Radul, and J. M. Siskind. 2018. “Automatic differentiation in machine learning: A survey.” J. Mach. Learn. Res. 18 (153): 1–43.
Bekele, Y. W. 2020. “Physics-informed deep learning for flow and deformation in poroelastic media.” Preprint, submitted October 29, 2020. https://arxiv.org/abs/2010.15426.
Bekele, Y. W. 2021. “Physics-informed deep learning for one-dimensional consolidation.” J. Rock Mech. Geotech. Eng. 13 (2): 420–430. https://doi.org/10.1016/j.jrmge.2020.09.005.
Brooks, R. H., and A. T. Corey. 1966. “Properties of porous media affecting fluid flow.” J. Irrig. Drain. Div. 92 (2): 61–88. https://doi.org/10.1061/JRCEA4.0000425.
Cai, S., Z. Mao, Z. Wang, M. Yin, and G. E. Karniadakis. 2022. “Physics-informed neural networks (PINNs) for fluid mechanics: A review.” Acta Mech. Sin. 37 (Jan): 1727–1738. https://doi.org/10.1007/s10409-021-01148-1.
Cai, S., Z. Wang, S. Wang, P. Perdikaris, and G. E. Karniadakis. 2021. “Physics-informed neural networks for heat transfer problems.” J. Heat Transfer 143 (6): 060801. https://doi.org/10.1115/1.4050542.
Coussy, O. 2004. Poromechanics. Chichester, UK: Wiley.
Dakshanamurthy, V., and D. Fredlund. 1981. “A mathematical model for predicting moisture flow in an unsaturated soil under hydraulic and temperature gradients.” Water Resour. Res. 17 (3): 714–722. https://doi.org/10.1029/WR017i003p00714.
Eivazi, H., M. Tahani, P. Schlatter, and R. Vinuesa. 2021. “Physics-informed neural networks for solving Reynolds-averaged Navier Stokes equations.” Preprint, submitted April 9, 2022. https://arxiv.org/abs/2107.10711.
Fraces, C. G., A. Papaioannou, and H. Tchelepi. 2020. “Physics informed deep learning for transport in porous media. Buckley leverett problem.” Preprint, submitted January 15, 2020. https://arxiv.org/abs/2001.05172.
Fraces, C. G., and H. Tchelepi. 2021. “Physics informed deep learning for flow and transport in porous media.” In Proc., SPE Reservoir Simulation Conf. Richardson, TX: OnePetro. https://doi.org/10.2118/203934-MS.
Fuks, O., and H. A. Tchelepi. 2020. “Limitations of physics informed machine learning for nonlinear two-phase transport in porous media.” J. Mach. Learn. Model. Comput. 1 (1): 19–37. https://doi.org/10.1615/jMachLearnModelComput.2020033905.
Gao, H., L. Sun, and J.-X. Wang. 2021. “Phygeonet: Physics-informed geometry-adaptive convolutional neural networks for solving parameterized steady-state pdes on irregular domain.” J. Comput. Phys. 428 (Mar): 110079. https://doi.org/10.1016/j.jcp.2020.110079.
Garipov, T. T., P. Tomin, R. Rin, D. V. Voskov, and H. A. Tchelepi. 2018. “Unified thermo-compositional-mechanical framework for reservoir simulation.” Comput. Geosci. 22 (4): 1039–1057. https://doi.org/10.1007/s10596-018-9737-5.
Gawin, D., B. A. Schrefler, and M. Galindo. 1996. “Thermo-hydro-mechanical analysis of partially saturated porous materials.” Eng. Comput. 13 (7): 113–143. https://doi.org/10.1108/02644409610151584.
Gelet, R., B. Loret, and N. Khalili. 2012. “A thermo-hydro-mechanical coupled model in local thermal non-equilibrium for fractured HDR reservoir with double porosity.” J. Geophys. Res.: Solid Earth 117 (B7): B07205. https://doi.org/10.1029/2012JB009161.
Goodfellow, I., Y. Bengio, and A. Courville. 2016. Deep learning. Cambridge, MA: MIT Press.
Haghighat, E., D. Amini, and R. Juanes. 2022. “Physics-informed neural network simulation of multiphase poroelasticity using stress-split sequential training.” Comput. Methods Appl. Mech. Eng. 397 (Jul): 115141. https://doi.org/10.1016/j.cma.2022.115141.
Haghighat, E., A. C. Bekar, E. Madenci, and R. Juanes. 2021a. “A nonlocal physics-informed deep learning framework using the peridynamic differential operator.” Comput. Methods Appl. Mech. Eng. 385 (Nov): 114012. https://doi.org/10.1016/j.cma.2021.114012.
Haghighat, E., and R. Juanes. 2021. “SciANN: A Keras/TensorFlow wrapper for scientific computations and physics-informed deep learning using artificial neural networks.” Comput. Methods Appl. Mech. Eng. 373 (Jan): 113552. https://doi.org/10.1016/j.cma.2020.113552.
Haghighat, E., M. Raissi, A. Moure, H. Gomez, and R. Juanes. 2021b. “A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics.” Comput. Methods Appl. Mech. Eng. 379 (Jun): 113741. https://doi.org/10.1016/j.cma.2021.113741.
Hennigh, O., S. Narasimhan, M. A. Nabian, A. Subramaniam, K. Tangsali, Z. Fang, M. Rietmann, W. Byeon, and S. Choudhry. 2021. “NVIDIA SimNet: An AI-accelerated multi-physics simulation framework.” In Proc., Int. Conf. on Computational Science, 447–461. Cham, Switzerland: Springer.
Hughes, T. J. 2012. The finite element method: Linear static and dynamic finite element analysis. New York: Dover.
Iranmanesh, M. A., A. Pak, and S. Samimi. 2018. “Non-isothermal simulation of the behavior of unsaturated soils using a novel EFG-based three dimensional model.” Comput. Geotech. 99 (Jul): 93–103. https://doi.org/10.1016/j.compgeo.2018.02.024.
Jagtap, A. D., and G. E. Karniadakis. 2020. “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. https://doi.org/10.4208/cicp.OA-2020-0164.
Jagtap, A. D., K. Kawaguchi, and G. E. Karniadakis. 2020. “Adaptive activation functions accelerate convergence in deep and physics-informed neural networks.” J. Comput. Phys. 404 (Mar): 109136. https://doi.org/10.1016/j.jcp.2019.109136.
Jha, B., and R. Juanes. 2014. “Coupled multiphase flow and poromechanics: A computational model of pore-pressure effects on fault slip and earthquake triggering.” Water Resour. Res. 50 (5): 3776–3808. https://doi.org/10.1002/2013WR015175.
Jin, X., S. Cai, H. Li, and G. E. Karniadakis. 2021. “NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations.” J. Comput. Phys. 426 (Feb): 109951. https://doi.org/10.1016/j.jcp.2020.109951.
Kadeethum, T., T. M. Jørgensen, and H. M. Nick. 2020. “Physics-informed neural networks for solving nonlinear diffusivity and biot’s equations.” PLoS One 15 (5): e0232683. https://doi.org/10.1371/journal.pone.0232683.
Karniadakis, G. E., I. G. Kevrekidis, L. Lu, P. Perdikaris, S. Wang, and L. Yang. 2021. “Physics-informed machine learning.” Nat. Rev. Phys. 3 (6): 422–440. https://doi.org/10.1038/s42254-021-00314-5.
Khoei, A. R., D. Amini, and S. M. S. Mortazavi. 2021. “Modeling non-isothermal two-phase fluid flow with phase change in deformable fractured porous media using extended finite element method.” Int. J. Numer. Methods Eng. 122 (16): 4378–4426. https://doi.org/10.1002/nme.6709.
Kim, J. 2018. “Unconditionally stable sequential schemes for all-way coupled thermoporomechanics: Undrained-adiabatic and extended fixed-stress splits.” Comput. Methods Appl. Mech. Eng. 341 (Nov): 93–112. https://doi.org/10.1016/j.cma.2018.06.030.
Kim, J., H. A. Tchelepi, and R. Juanes. 2011a. “Stability and convergence of sequential methods for coupled flow and geomechanics: Drained and undrained splits.” Comput. Methods Appl. Mech. Eng. 200 (23–24): 2094–2116. https://doi.org/10.1016/j.cma.2011.02.011.
Kim, J., H. A. Tchelepi, and R. Juanes. 2011b. “Stability and convergence of sequential methods for coupled flow and geomechanics: Fixed-stress and fixed-strain splits.” Comput. Methods Appl. Mech. Eng. 200 (13–16): 1591–1606. https://doi.org/10.1016/j.cma.2010.12.022.
Kim, J., H. A. Tchelepi, and R. Juanes. 2013. “Rigorous coupling of geomechanics and multiphase flow with strong capillarity.” SPE J. 18 (6): 1123–1139. https://doi.org/10.2118/141268-PA.
Kingma, D. P., and J. Ba. 2014. “Adam: A method for stochastic optimization.” Preprint, submitted December 22, 2014. https://arxiv.org/abs/1412.6980.
Laubscher, R. 2021. “Simulation of multi-species flow and heat transfer using physics-informed neural networks.” Phys. Fluids 33 (8): 087101. https://doi.org/10.1063/5.0058529.
Lewis, R., and B. Schrefler. 1998. The finite element method in the static and dynamic deformation and consolidation of porous media. Chichester, UK: Wiley.
Mao, Z., A. D. Jagtap, and G. E. Karniadakis. 2020. “Physics-informed neural networks for high-speed flows.” Comput. Methods Appl. Mech. Eng. 360 (Mar): 112789. https://doi.org/10.1016/j.cma.2019.112789.
Nabian, M. A., R. J. Gladstone, and H. Meidani. 2021. “Efficient training of physics-informed neural networks via importance sampling.” Comput.-Aided Civ. Infrastruct. Eng. 36 (8): 962–977. https://doi.org/10.1111/mice.12685.
Niaki, S. A., E. Haghighat, T. Campbell, A. Poursartip, and R. Vaziri. 2021. “Physics-informed neural network for modelling the thermochemical curing process of composite-tool systems during manufacture.” Comput. Methods Appl. Mech. Eng. 384 (Oct): 113959. https://doi.org/10.1016/j.cma.2021.113959.
Nocedal, J., and S. Wright. 2006. Numerical optimization. New York: Springer.
Noorishad, J., C. Tsang, and P. Witherspoon. 1984. “Coupled thermal-hydraulic-mechanical phenomena in saturated fractured porous rocks: Numerical approach.” J. Geophys. Res.: Solid Earth 89 (B12): 10365–10373. https://doi.org/10.1029/JB089iB12p10365.
O’Sullivan, M. J., K. Pruess, and M. J. Lippmann. 2001. “State of the art of geothermal reservoir simulation.” Geothermics 30 (4): 395–429. https://doi.org/10.1016/S0375-6505(01)00005-0.
Pandey, S., A. Chaudhuri, and S. Kelkar. 2017. “A coupled thermo-hydro-mechanical modeling of fracture aperture alteration and reservoir deformation during heat extraction from a geothermal reservoir.” Geothermics 65 (Jan): 17–31. https://doi.org/10.1016/j.geothermics.2016.08.006.
Pao, W. K., R. W. Lewis, and I. Masters. 2001. “A fully coupled hydro-thermo-poro-mechanical model for black oil reservoir simulation.” Int. J. Numer. Anal. Methods Geomech. 25 (12): 1229–1256. https://doi.org/10.1002/nag.174.
Praditia, T., R. Helmig, and H. Hajibeygi. 2018. “Multiscale formulation for coupled flow-heat equations arising from single-phase flow in fractured geothermal reservoirs.” Comput. Geosci. 22 (5): 1305–1322. https://doi.org/10.1007/s10596-018-9754-4.
Raissi, M., P. Perdikaris, and G. E. Karniadakis. 2019. “Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations.” J. Comput. Phys. 378 (Feb): 686–707. https://doi.org/10.1016/j.jcp.2018.10.045.
Ranade, R., C. Hill, and J. Pathak. 2021. “DiscretizationNet: A machine-learning based solver for Navier–Stokes equations using finite volume discretization.” Comput. Methods Appl. Mech. Eng. 378 (May): 113722. https://doi.org/10.1016/j.cma.2021.113722.
Rao, C., H. Sun, and Y. Liu. 2020. “Physics-informed deep learning for incompressible laminar flows.” Theor. Appl. Mech. Lett. 10 (3): 207–212. https://doi.org/10.1016/j.taml.2020.01.039.
Rao, C., H. Sun, and Y. Liu. 2021. “Physics-informed deep learning for computational elastodynamics without labeled data.” J. Eng. Mech. 147 (8): 04021043. https://doi.org/10.1061/(ASCE)EM.1943-7889.0001947.
Reyes, B., A. A. Howard, P. Perdikaris, and A. M. Tartakovsky. 2021. “Learning unknown physics of non-Newtonian fluids.” Phys. Rev. Fluids 6 (7): 073301. https://doi.org/10.1103/PhysRevFluids.6.073301.
Rutqvist, J., D. Barr, J. T. Birkholzer, K. Fujisaki, O. Kolditz, Q.-S. Liu, T. Fujita, W. Wang, and C.-Y. Zhang. 2009. “A comparative simulation study of coupled THM processes and their effect on fractured rock permeability around nuclear waste repositories.” Environ. Geol. 57 (6): 1347–1360. https://doi.org/10.1007/s00254-008-1552-1.
Rutqvist, J., L. Börgesson, M. Chijimatsu, A. Kobayashi, L. Jing, T. Nguyen, J. Noorishad, and C.-F. Tsang. 2001. “Thermohydromechanics of partially saturated geological media: Governing equations and formulation of four finite element models.” Int. J. Rock Mech. Min. Sci. 38 (1): 105–127. https://doi.org/10.1016/S1365-1609(00)00068-X.
Schrefler, B., L. Simoni, and E. Turska. 1997. “Standard staggered and staggered Newton schemes in thermo-hydro-mechanical problems.” Comput. Methods Appl. Mech. Eng. 144 (1–2): 93–109. https://doi.org/10.1016/S0045-7825(96)01170-X.
Schrefler, B. A., X. Zhan, and L. Simoni. 1995. “A coupled model for water flow, airflow and heat flow in deformable porous media.” Int. J. Numer. Methods Heat Fluid Flow 5 (6): 531–547. https://doi.org/10.1108/EUM0000000004077.
Wang, S., H. Wang, and P. Perdikaris. 2021. “On the eigenvector bias of Fourier feature networks: From regression to solving multi-scale PDEs with physics-informed neural networks.” Comput. Methods Appl. Mech. Eng. 384 (Oct): 113938. https://doi.org/10.1016/j.cma.2021.113938.
Wang, S., X. Yu, and P. Perdikaris. 2022. “When and why PINNs fail to train: A neural tangent kernel perspective.” J. Comput. Phys. 449 (Jan): 110768. https://doi.org/10.1016/j.jcp.2021.110768.
Wang, W., G. Kosakowski, and O. Kolditz. 2009. “A parallel finite element scheme for thermo-hydro-mechanical (THM) coupled problems in porous media.” Comput. Geosci. 35 (8): 1631–1641. https://doi.org/10.1016/j.cageo.2008.07.007.
Weinan, E., and B. Yu. 2018. “The deep Ritz method: A deep learning-based numerical algorithm for solving variational problems.” Commun. Math. Stat. 6 (1): 1–12. https://doi.org/10.1007/s40304-018-0127-z.
Wu, J.-L., H. Xiao, and E. Paterson. 2018. “Physics-informed machine learning approach for augmenting turbulence models: A comprehensive framework.” Phys. Rev. Fluids 3 (7): 074602. https://doi.org/10.1103/PhysRevFluids.3.074602.
Zienkiewicz, O. C., R. L. Taylor, P. Nithiarasu, and J. Zhu. 1977. Vol. 3 of The finite element method. London: McGraw-Hill.
Information & Authors
Information
Published In
Journal of Engineering Mechanics
Volume 148 • Issue 11 • November 2022
Copyright
© 2022 American Society of Civil Engineers.
History
Received: Mar 6, 2022
Accepted: Jun 29, 2022
Published online: Sep 15, 2022
Published in print: Nov 1, 2022
Discussion open until: Feb 15, 2023
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Cited by
- Pin Zhang, Zhen-Yu Yin, Brian Sheil, A physics-informed data-driven approach for consolidation analysis, Géotechnique, 10.1680/jgeot.22.00046, (1-12), (2023).
- Shuyi Zhou, Fanghua Xu, Ruizi Shi, Whitecap Fraction Parameterization and Understanding with Deep Neural Network, Remote Sensing, 10.3390/rs15010241, 15, 1, (241), (2022).
- Haotian Chen, Christopher Batchelor-McAuley, Enno Kätelhön, Joseph Elliott, Richard G. Compton, A Critical Evaluation of Using Physics-Informed Neural Networks for Simulating Voltammetry: Strengths, Weaknesses and Best Practices, Journal of Electroanalytical Chemistry, 10.1016/j.jelechem.2022.116918, 925, (116918), (2022).