Abstract
This paper presents the fundamentals of Physics Informed Neural Networks (PINNs) and reviews literature on the methodology and application of PINNs. PINNs are universal approximators that integrates physical laws that can be described by partial differential equations (PDEs) and given data, in the learning process. The formulations of PINNs are first presented in an example of linear elasticity problem. Then, the characteristics of PINNs are compared with that of conventional numerical analysis approach. A review of the literature on PINNs for solving not only forward, but also inverse problems is discussed. In addition, special attention is given to the employment of PINNs in multiscale analysis.
Similar content being viewed by others
References
M. Raissi, P. Perdikaris, G.E. Karniadakis, 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://doi.org/10.1016/j.jcp.2018.10.045
G.E. Karniadakis, I.G. Kevrekidis, L. Lu, P. Perdikaris, S. Wang, L. Yang, Physics-informed machine learning. Nat. Rev. Phys. 3(6), 422–440 (2021). https://doi.org/10.1038/s42254-021-00314-5
O. Fuks, H.A. Tchelepi, Limitations of physics informed machine learning for nonlinear two-phase transport in porous media. J. Mach. Learn. Model. Comput. 1(1), 19–37 (2020). https://doi.org/10.1615/JMachLearnModelComput.2020033905
T.G. Grossmann, U.J. Komorowska, J. Latz, C.-B. Schönlieb, Can physics-informed neural networks beat the finite element method? arXiv:2302.04107 (2023)
M.-S. Go, J.H. Lim, S. Lee, Physics-informed neural network-based surrogate model for a virtual thermal sensor with real-time simulation. Int. J. Heat Mass Transf. 214, 124392 (2023). https://doi.org/10.1016/j.ijheatmasstransfer.2023.124392
M. Raissi, A. Yazdani, G.E. Karniadakis, Hidden fluid mechanics: learning velocity and pressure fields from flow visualizations. Science 367(6481), 1026–1030 (2020). https://doi.org/10.1126/science.aaw4741
E. Zhang, M. Dao, G.E. Karniadakis, S. Suresh, Analyses of internal structures and defects in materials using physics-informed neural networks. Sci. Adv. 8(7), 0644 (2022). https://doi.org/10.1126/sciadv.abk0644
A. Kamali, M. Sarabian, K. Laksari, Elasticity imaging using physics-informed neural networks: spatial discovery of elastic modulus and Poisson’s ratio. Available at SSRN 4203110
L. Lu, X. Meng, Z. Mao, G.E. Karniadakis, Deepxde: a deep learning library for solving differential equations. SIAM Rev. 63(1), 208–228 (2021). https://doi.org/10.1137/19m1274067
E. Kharazmi, Z. Zhang, G.E.M. Karniadakis, HP-VPINNS: variational physics-informed neural networks with domain decomposition. Comput. Methods Appl. Mech. Eng. 374, 113547 (2021). https://doi.org/10.1016/j.cma.2020.113547
E. Samaniego, C. Anitescu, S. Goswami, V.M. Nguyen-Thanh, H. Guo, K. Hamdia, X. Zhuang, T. Rabczuk, An energy approach to the solution of partial differential equations in computational mechanics via machine learning: Concepts, implementation and applications. Comput. Methods Appl. Mech. Eng. 362, 112790 (2020). https://doi.org/10.1016/j.cma.2019.112790
C. Rao, H. Sun, Y. Liu, Hard encoding of physics for learning spatiotemporal dynamics. arXiv:2105.00557 (2021)
N. Geneva, N. Zabaras, Modeling the dynamics of PDE systems with physics-constrained deep auto-regressive networks. J. Comput. Phys. 403, 109056 (2020). https://doi.org/10.1016/j.jcp.2019.109056
H. Gao, L. Sun, J.-X. Wang, Phygeonet: physics-informed geometry-adaptive convolutional neural networks for solving parameterized steady-state PDES on irregular domain. J. Comput. Phys. 428, 110079 (2021). https://doi.org/10.1016/j.jcp.2020.110079
N. Wandel, M. Weinmann, M. Neidlin, R. Klein, Spline-PINN: approaching PDES without data using fast, physics-informed hermite-spline CNNS. Proc. AAAI Conf. Artif. Intell. 36(8), 8529–8538 (2022). https://doi.org/10.1609/aaai.v36i8.20830
S.F. Wang, Y.J. Teng, P. Perdikaris, Understanding and mitigating gradient flow pathologies in physics-informed neural networks. SIAM J. Sci. Comput. 43(5), 3055–3081 (2021). https://doi.org/10.1137/20m1318043. (Xg7ux Times Cited:214 Cited References Count:57)
L. McClenny, U. Braga-Neto, Self-adaptive physics-informed neural networks using a soft attention mechanism. arXiv:2009.04544 (2020)
P.-H. Chiu, J.C. Wong, C. Ooi, M.H. Dao, Y.-S. Ong, Can-PINN: a fast physics-informed neural network based on coupled-automatic-numerical differentiation method. Comput. Methods Appl. Mech. Eng. 395, 114909 (2022). https://doi.org/10.1016/j.cma.2022.114909
L. Yuan, Y.-Q. Ni, X.-Y. Deng, S. Hao, A-PINN: auxiliary physics informed neural networks for forward and inverse problems of nonlinear integro-differential equations. J. Comput. Phys. 462, 111260 (2022). https://doi.org/10.1016/j.jcp.2022.111260
E. Haghighat, A.C. Bekar, E. Madenci, R. Juanes, A nonlocal physics-informed deep learning framework using the peridynamic differential operator. Comput. Methods Appl. Mech. Eng. 385, 114012 (2021). https://doi.org/10.1016/j.cma.2021.114012
A.D. Jagtap, E. Kharazmi, G.E. Karniadakis, 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://doi.org/10.1016/j.cma.2020.113028
A.D. Jagtap, G.E. Karniadakis, 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, 2002–2041 (2020). https://doi.org/10.4208/cicp.OA-2020-0164
V. Dwivedi, N. Parashar, B. Srinivasan, Distributed learning machines for solving forward and inverse problems in partial differential equations. Neurocomputing 420, 299–316 (2021). https://doi.org/10.1016/j.neucom.2020.09.006
S. Goswami, A. Bora, Y. Yu, G.E. Karniadakis, in Physics-Informed Deep Neural Operator Networks ed. by T. Rabczuk, K.-J. Bathe (Springer, Cham, 2023), pp. 219–254. https://doi.org/10.1007/978-3-031-36644-4_6
L. Lu, P. Jin, G.E. Karniadakis, Deeponet: learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators. arXiv:1910.03193 (2019)
L. Lu, P. Jin, G. Pang, Z. Zhang, G.E. Karniadakis, Learning nonlinear operators via deeponet based on the universal approximation theorem of operators. Nat. Mach. Intell. 3(3), 218–229 (2021). https://doi.org/10.1038/s42256-021-00302-5
S. Wang, H. Wang, P. Perdikaris, Learning the solution operator of parametric partial differential equations with physics-informed deeponets. Sci. Adv. 7(40), 8605 (2021). https://doi.org/10.1126/sciadv.abi8605
Z. Li, N. Kovachki, K. Azizzadenesheli, B. Liu, K. Bhattacharya, A. Stuart, A. Anandkumar, Fourier neural operator for parametric partial differential equations. in ICLR 2021: The Ninth International Conference on Learning Representations (2021)
Z. Li, H. Zheng, N. Kovachki, D. Jin, H. Chen, B. Liu, K. Azizzadenesheli, A. Anandkumar, Physics-informed neural operator for learning partial differential equations. arXiv:2111.03794 (2021)
R. Ranade, C. Hill, J. Pathak, Discretizationnet: a machine-learning based solver for Navier–Stokes equations using finite volume discretization. Comput. Methods Appl. Mech. Eng. 378, 113722 (2021). https://doi.org/10.1016/j.cma.2021.113722
J. Oldenburg, F. Borowski, A. Öner, K.-P. Schmitz, M. Stiehm, Geometry aware physics informed neural network surrogate for solving Navier–Stokes equation (GAPINN). Adv. Model. Simul. Eng. Sci. 9(1), 8 (2022). https://doi.org/10.1186/s40323-022-00221-z
C. Cheng, G.-T. Zhang, Deep learning method based on physics informed neural network with resnet block for solving fluid flow problems. Water 13(4), 423 (2021)
M. Mahmoudabadbozchelou, G.E. Karniadakis, S. Jamali, NN-PINNS: non-Newtonian physics-informed neural networks for complex fluid modeling. Soft Matter 18(1), 172–185 (2022)
M. Mahmoudabadbozchelou, S. Jamali, Rheology-informed neural networks (RHINNS) for forward and inverse metamodelling of complex fluids. Sci. Rep. 11(1), 12015 (2021). https://doi.org/10.1038/s41598-021-91518-3
S. Thakur, M. Raissi, A.M. Ardekani, Viscoelasticnet: a physics informed neural network framework for stress discovery and model selection. arXiv:2209.06972 (2022)
E. Haghighat, D. Amini, R. Juanes, Physics-informed neural network simulation of multiphase poroelasticity using stress-split sequential training. Comput. Methods Appl. Mech. Eng. 397, 115141 (2022). https://doi.org/10.1016/j.cma.2022.115141
M.M. Almajid, M.O. Abu-Al-Saud, Prediction of porous media fluid flow using physics informed neural networks. J. Petrol. Sci. Eng. 208, 109205 (2022). https://doi.org/10.1016/j.petrol.2021.109205
H. Wessels, C. Weißenfels, P. Wriggers, The neural particle method—an updated Lagrangian physics informed neural network for computational fluid dynamics. Comput. Methods Appl. Mech. Eng. 368, 113127 (2020). https://doi.org/10.1016/j.cma.2020.113127
T. Zhang, B. Dey, P. Kakkar, A. Dasgupta, A. Chakraborty, Frequency-compensated PINNS for fluid-dynamic design problems. arXiv:2011.01456 (2020)
E. Haghighat, M. Raissi, A. Moure, H. Gomez, R. Juanes, A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics. Comput. Methods Appl. Mech. Eng. 379, 113741 (2021). https://doi.org/10.1016/j.cma.2021.113741
C. Rao, H. Sun, Y. Liu, Physics-informed deep learning for computational elastodynamics without labeled data. J. Eng. Mech. 147(8), 04021043 (2021). https://doi.org/10.1061/(ASCE)EM.1943-7889.0001947
G.K. Yadav, S. Natarajan, B. Srinivasan, Distributed PINN for linear elasticity—a unified approach for smooth, singular, compressible and incompressible media. Int. J. Comput. Methods 19(08), 2142008 (2022). https://doi.org/10.1142/s0219876221420081
A. Henkes, H. Wessels, R. Mahnken, Physics informed neural networks for continuum micromechanics. Comput. Methods Appl. Mech. Eng. 393, 114790 (2022). https://doi.org/10.1016/j.cma.2022.114790
D.W. Abueidda, S. Koric, R.A. Al-Rub, C.M. Parrott, K.A. James, N.A. Sobh, A deep learning energy method for hyperelasticity and viscoelasticity. Eur. J. Mech. A. Solids 95, 104639 (2022). https://doi.org/10.1016/j.euromechsol.2022.104639
J.N. Fuhg, N. Bouklas, The mixed deep energy method for resolving concentration features in finite strain hyperelasticity. J. Comput. Phys. 451, 110839 (2022). https://doi.org/10.1016/j.jcp.2021.110839
D.W. Abueidda, S. Koric, E. Guleryuz, N.A. Sobh, Enhanced physics-informed neural networks for hyperelasticity. Int. J. Numer. Meth. Eng. 124(7), 1585–1601 (2023). https://doi.org/10.1002/nme.7176
S. Rezaei, A. Harandi, A. Moeineddin, B.-X. Xu, S. Reese, A mixed formulation for physics-informed neural networks as a potential solver for engineering problems in heterogeneous domains: comparison with finite element method. Comput. Methods Appl. Mech. Eng. 401, 115616 (2022). https://doi.org/10.1016/j.cma.2022.115616
J. Bai, T. Rabczuk, A. Gupta, L. Alzubaidi, Y. Gu, A physics-informed neural network technique based on a modified loss function for computational 2d and 3d solid mechanics. Comput. Mech. (2022). https://doi.org/10.1007/s00466-022-02252-0
C.A. Yan, R. Vescovini, L. Dozio, A framework based on physics-informed neural networks and extreme learning for the analysis of composite structures. Comput. Struct. 265, 106761 (2022). https://doi.org/10.1016/j.compstruc.2022.106761
G.-B. Huang, Q.-Y. Zhu, C.-K. Siew, Extreme learning machine: theory and applications. Neurocomputing 70(1), 489–501 (2006). https://doi.org/10.1016/j.neucom.2005.12.126
S. Goswami, M. Yin, Y. Yu, G.E. Karniadakis, A physics-informed variational deeponet for predicting crack path in quasi-brittle materials. Comput. Methods Appl. Mech. Eng. 391, 114587 (2022). https://doi.org/10.1016/j.cma.2022.114587
Z. Zhang, G.X. Gu, Physics-informed deep learning for digital materials. Theor. Appl. Mech. Lett. 11(1), 100220 (2021). https://doi.org/10.1016/j.taml.2021.100220
L.H. Song, C. Wang, J.S. Fan, H.M. Lu, Elastic structural analysis based on graph neural network without labeled data. Comput.-Aid. Civil Infrastruct. Eng. 38(10), 1307–1323 (2023). https://doi.org/10.1111/mice.12944. (I4gv2 Times Cited:1 Cited References Count:60)
R. Arora, Physrnet: physics informed super-resolution network for application in computational solid mechanics. in 2022 IEEE/ACM International Workshop on Artificial Intelligence and Machine Learning for Scientific Applications (AI4S), 13–18. (2022). https://doi.org/10.1109/AI4S56813.2022.00008
C.M. Hamel, K.N. Long, S.L. Kramer, Calibrating constitutive models with full-field data via physics informed neural networks. arXiv:2203.16577 (2022)
T. Kadeethum, T.M. Jørgensen, H.M. Nick, Physics-informed neural networks for solving inverse problems of nonlinear biot’s equations: batch training. in 54th US Rock Mechanics/Geomechanics Symposium. OnePetro
Z. Mao, A.D. Jagtap, G.E. Karniadakis, Physics-informed neural networks for high-speed flows. Comput. Methods Appl. Mech. Eng. 360, 112789 (2020). https://doi.org/10.1016/j.cma.2019.112789
A.M. Tartakovsky, C.O. Marrero, P. Perdikaris, G.D. Tartakovsky, D. Barajas-Solano, Learning parameters and constitutive relationships with physics informed deep neural networks. arXiv:1808.03398 (2018)
Y. Deng, C. Chen, Q. Wang, X. Li, Z. Fan, Y. Li, Modeling a typical non-uniform deformation of materials using physics-informed deep learning: applications to forward and inverse problems. Appl. Sci. 13(7), 4539 (2023)
K. Shukla, A.D. Jagtap, J.L. Blackshire, D. Sparkman, G.E. Karniadakis, A physics-informed neural network for quantifying the microstructural properties of polycrystalline nickel using ultrasound data: a promising approach for solving inverse problems. IEEE Signal Process. Mag. 39(1), 68–77 (2022). https://doi.org/10.1109/MSP.2021.3118904
E. Zhang, M. Yin, G.E. Karniadakis, Physics-informed neural networks for nonhomogeneous material identification in elasticity imaging. arXiv:2009.04525 (2020)
Y. Wei, Q. Serra, G. Lubineau, E. Florentin, Coupling physics-informed neural networks and constitutive relation error concept to solve a parameter identification problem. Comput. Struct. 283, 107054 (2023). https://doi.org/10.1016/j.compstruc.2023.107054
C.-T. Chen, G.X. Gu, Physics-informed deep-learning for elasticity: forward, inverse, and mixed problems. Adv. Sci. (2023). https://doi.org/10.1002/advs.202300439
S. Mowlavi, K. Kamrin, Topology optimization with physics-informed neural networks: application to noninvasive detection of hidden geometries. arXiv:2303.09280 (2023)
Z. Fang, J. Zhan, Deep physical informed neural networks for metamaterial design. IEEE Access 8, 24506–24513 (2020). https://doi.org/10.1109/ACCESS.2019.2963375
J. Zehnder, Y. Li, S. Coros, B. Thomaszewski, Ntopo: mesh-free topology optimization using implicit neural representations. Adv. Neural. Inf. Process. Syst. 34, 10368–10381 (2021)
A. Joglekar, H. Chen, L.B. Kara, DMF-TONN: direct mesh-free topology optimization using neural networks. arXiv:2305.04107 (2023)
L. Lu, R. Pestourie, W. Yao, Z. Wang, F. Verdugo, S.G. Johnson, Physics-informed neural networks with hard constraints for inverse design. SIAM J. Sci. Comput. 43(6), 1105–1132 (2021). https://doi.org/10.1137/21m1397908
L. Lu, R. Pestourie, S.G. Johnson, G. Romano, Multifidelity deep neural operators for efficient learning of partial differential equations with application to fast inverse design of nanoscale heat transport. Phys. Rev. Res. 4(2), 023210 (2022). https://doi.org/10.1103/PhysRevResearch.4.023210. (PRRESEARCH)
K. Wu, X.-B. Yan, S. Jin, Z. Ma, Capturing the diffusive behavior of the multiscale linear transport equations by asymptotic-preserving convolutional deeponets. Comput. Methods Appl. Mech. Eng. 418, 116531 (2024). https://doi.org/10.1016/j.cma.2023.116531
S. Cai, Z. Wang, L. Lu, T.A. Zaki, G.E. Karniadakis, Deepm &mnet: Inferring the electroconvection multiphysics fields based on operator approximation by neural networks. J. Comput. Phys. 436, 110296 (2021). https://doi.org/10.1016/j.jcp.2021.110296
Z. Mao, L. Lu, O. Marxen, T.A. Zaki, G.E. Karniadakis, Deepm &mnet for hypersonics: predicting the coupled flow and finite-rate chemistry behind a normal shock using neural-network approximation of operators. J. Comput. Phys. 447, 110698 (2021). https://doi.org/10.1016/j.jcp.2021.110698
C. Lin, Z. Li, L. Lu, S. Cai, M. Maxey, G.E. Karniadakis, Operator learning for predicting multiscale bubble growth dynamics. J. Chem. Phys. (2021). https://doi.org/10.1063/5.0041203
L. Liu, W. Cai, Multiscale deeponet for nonlinear operators in oscillatory function spaces for building seismic wave responses. arXiv:2111.04860 (2021)
W.T. Leung, G. Lin, Z. Zhang, Nh-pinn: neural homogenization-based physics-informed neural network for multiscale problems. J. Comput. Phys. 470, 111539 (2022). https://doi.org/10.1016/j.jcp.2022.111539
J. Han, Y. Lee, A neural network approach for homogenization of multiscale problems. Multiscale Model. Simul. 21(2), 716–734 (2023). https://doi.org/10.1137/22m1500903
Z. Zhang, C. Moya, W.T. Leung, G. Lin, H. Schaeffer, Bayesian deep operator learning for homogenized to fine-scale maps for multiscale PDE. arXiv:2308.14188 (2023)
M. Yin, E. Zhang, Y. Yu, G.E. Karniadakis, Interfacing finite elements with deep neural operators for fast multiscale modeling of mechanics problems. Comput. Methods Appl. Mech. Eng. 402, 115027 (2022). https://doi.org/10.1016/j.cma.2022.115027
S.E. Ahmed, P. Stinis, A multifidelity deep operator network approach to closure for multiscale systems. Comput. Methods Appl. Mech. Eng. 414, 116161 (2023). https://doi.org/10.1016/j.cma.2023.116161
C. Lin, M. Maxey, Z. Li, G.E. Karniadakis, A seamless multiscale operator neural network for inferring bubble dynamics. J. Fluid Mech. 929, 18 (2021). https://doi.org/10.1017/jfm.2021.866
R. Wang, K. Kashinath, M. Mustafa, A. Albert, R. Yu, Towards physics-informed deep learning for turbulent flow prediction. Assoc. Comput. Mach. (2020). https://doi.org/10.1145/3394486.3403198
E.J. Hall, S. Taverniers, M.A. Katsoulakis, D.M. Tartakovsky, Ginns: graph-informed neural networks for multiscale physics. J. Comput. Phys. 433, 110192 (2021). https://doi.org/10.1016/j.jcp.2021.110192
J. Wu, J. Jiang, Q. Chen, G. Chatzigeorgiou, F. Meraghni, Deep homogenization networks for elastic heterogeneous materials with two- and three-dimensional periodicity. Int. J. Solids Struct. 284, 112521 (2023). https://doi.org/10.1016/j.ijsolstr.2023.112521
Y. Zhu, N. Zabaras, P.-S. Koutsourelakis, P. Perdikaris, Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data. J. Comput. Phys. 394, 56–81 (2019). https://doi.org/10.1016/j.jcp.2019.05.024
Q. Lou, X. Meng, G.E. Karniadakis, Physics-informed neural networks for solving forward and inverse flow problems via the Boltzmann-Bgk formulation. J. Comput. Phys. 447, 110676 (2021). https://doi.org/10.1016/j.jcp.2021.110676
Acknowledgements
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) (NRF-2023R1A2C2003758).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of Interest
The authors declare that they have no known competing financial interests that are directly or indirectly related to the work submitted for publication.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Kim, D., Lee, J. A Review of Physics Informed Neural Networks for Multiscale Analysis and Inverse Problems. Multiscale Sci. Eng. (2024). https://doi.org/10.1007/s42493-024-00106-w
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s42493-024-00106-w