Physics-Informed Machine Learning and Uncertainty Quantification for Mechanics of Heterogeneous Materials
Source abstract
Abstract
In this work, a model based on the Physics - Informed Neural Networks (PINNs) for solving elastic deformation of heterogeneous solids and associated Uncertainty Quantification (UQ) is presented. For the present study, the PINNs framework - Modulus developed by Nvidia is utilized, wherein we implement a module for mechanics of heterogeneous solids. We use PINNs to approximate momentum balance by assuming isotropic linear elastic constitutive behavior against a loss function. Along with governing equations, the associated initial / boundary conditions also softly participate in the loss function. Solids where the heterogeneity manifests as voids (low elastic modulus regions) and fibers (high elastic modulus regions) in a matrix are analyzed, and the results are validated against solutions obtained from a commercial Finite Element (FE) analysis package. The present study also reveals that PINNs can capture the stress jumps precisely at the material interfaces. Additionally, the present study explores the advantages associated with the surrogate features in PINNs via the variation in geometry and material properties. The presented UQ studies suggest that the mean and standard deviation of the PINNs solution are in good agreement with Monte Carlo FE results. The effective Young's modulus predicted by PINNs for single representative void and single fiber composites compare very well against the ones predicted by FE, which establishes the PINNs formulation as an efficient homogenization tool.
Source: openaire · Quality: candidate unverified · Source record ↗
Research context
OpenAlex topics: Model Reduction and Neural Networks · Probabilistic and Robust Engineering Design · Non-Destructive Testing Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (70)
Show source-provided references
1. Mehdikhani M, Gorbatikh L, Verpoest I, Lomov SV (2019) Voids in fiber-reinforced polymer composites: a review on their formation, characteristics, and effects on mechanical performance. J Compos Mater 53(12):1579–1669. https://doi.org/10.1177/002199831877215210.1177/0021998318772152 ↗
2. Huang H, Talreja R (2005) Effects of void geometry on elastic properties of unidirectional fiber reinforced composites. Compos Sci Technol 65(13):1964–198110.1016/j.compscitech.2005.02.019 ↗
3. Malakooti MH, Sodano HA (2013) Multi-inclusion modeling of multiphase piezoelectric composites. Compos B Eng 47:181–18910.1016/j.compositesb.2012.10.034 ↗
4. Beniwal A, Dadhich R, Alankar A (2019) Deep learning based predictive modeling for structure-property linkages. Materialia 8:10043510.1016/j.mtla.2019.100435 ↗
5. Revi V, Kasodariya S, Talapatra A, Pilania G, Alankar A (2021) Machine learning elastic constants of multi-component alloys. Comput Mater Sci 198:11067110.1016/j.commatsci.2021.110671 ↗
6. Liu X, Tian S, Tao F, Yu W (2021) A review of artificial neural networks in the constitutive modeling of composite materials. Compos B Eng 224:109152
7. Wang S, Teng Y, Perdikaris P (2021) Understanding and mitigating gradient flow pathologies in physics-informed neural networks. SIAM J Sci Comput 43(5):A3055–A308110.1137/20M1318043 ↗
8. Abueidda DW, Koric S, Sobh NA (2020) Topology optimization of 2d structures with nonlinearities using deep learning. Comput Struct 237:10628310.1016/j.compstruc.2020.106283 ↗
9. Goodfellow I, Bengio Y, Courville A (2016) Deep learning, MIT Press, http://www.deeplearningbook.org
10. Bhutada A, Kumar S, Gunasegaram D, Alankar A (2021) Machine learning based methods for obtaining correlations between microstructures and thermal stresses. Metals 11(8):116710.3390/met11081167 ↗
11. Lagaris IE, Likas A, Fotiadis DI (1998) Artificial neural networks for solving ordinary and partial differential equations. IEEE Trans Neural Netw 9(5):987–100010.1109/72.712178 ↗
12. Raissi M, Perdikaris P, Karniadakis GE (2017) Physics informed deep learning (part i): Data-driven solutions of nonlinear partial differential equations, arXiv preprint arXiv:1711.10561
13. Hornik K (1991) Approximation capabilities of multilayer feedforward networks. Neural Netw 4(2):251–25710.1016/0893-6080(91)90009-T ↗
14. Cybenko G (1989) Approximation by superpositions of a sigmoidal function. Math Control Signals Syst 2(4):303–31410.1007/BF02551274 ↗
15. Rudy S, Alla A, Brunton SL, Kutz JN (2019) Data-driven identification of parametric partial differential equations. SIAM J Appl Dyn Syst 18(2):643–66010.1137/18M1191944 ↗
16. Wight CL, Zhao J (2020) Solving allen-cahn and cahn-hilliard equations using the adaptive physics informed neural networks, arXiv preprint arXiv:2007.04542
17. Nguyen-Thanh VM, Zhuang X, Rabczuk T (2020) A deep energy method for finite deformation hyperelasticity. Eur J Mech A/Solids 80:103874
18. Haghighat E, Raissi M, Moure A, Gomez H, Juanes R (2021) A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics. Comput Methods Appl Mech Eng 379:11374110.1016/j.cma.2021.113741 ↗
19. Yang X, Zafar S, Wang J-X, Xiao H (2019) Predictive large-eddy-simulation wall modeling via physics-informed neural networks. Phys Rev Fluids 4(3):03460210.1103/PhysRevFluids.4.034602 ↗
20. Jin X, Cai S, Li H, Karniadakis GE (2021) Nsfnets (navier-stokes flow nets): Physics-informed neural networks for the incompressible navier-stokes equations. J Comput Phys 426:109951
21. Mahmoudabadbozchelou M, Karniadakis GE, Jamali S (2022) nn-pinns: Non-newtonian physics-informed neural networks for complex fluid modeling, Soft Matter10.1039/D1SM01298C ↗
22. Almqvist A (2021) Fundamentals of physics-informed neural networks applied to solve the reynolds boundary value problem. Lubricants 9(8):8210.3390/lubricants9080082 ↗
23. Cai S, Wang Z, Wang S, Perdikaris P, Karniadakis GE (2021) Physics-informed neural networks for heat transfer problems. J Heat Transf 143(6):060801
24. Mishra S, Molinaro R (2021) Physics informed neural networks for simulating radiative transfer. J Quant Spectrosc Radiat Transf 270:10770510.1016/j.jqsrt.2021.107705 ↗
25. Niaki SA, Haghighat E, Campbell T, Poursartip A, Vaziri R (2021) Physics-informed neural network for modelling the thermochemical curing process of composite-tool systems during manufacture. Comput Methods Appl Mech Eng 384:113959
26. Rocha F, Deparis S, Antolin P, Buffa A (2021) Deepbnd: a machine learning approach to enhance multiscale solid mechanics, arXiv preprint arXiv:2110.11141
27. Raissi M, Perdikaris P, Karniadakis GE (2017) Physics informed deep learning (part ii): Data-driven discovery of nonlinear partial differential equations, arXiv preprint arXiv:1711.10566
28. Basdevant C, Deville M, Haldenwang P, Lacroix J, Ouazzani J, Peyret R, Orlandi P, Patera A (1986) Spectral and finite difference solutions of the burgers equation. Comput Fluids 14(1):23–4110.1016/0045-7930(86)90036-8 ↗
29. Schrödinger E (1926) Quantisierung als eigenwertproblem. Ann Phys 385(13):437–49010.1002/andp.19263851302 ↗
30. Allen SM, Cahn JW (1979) A microscopic theory for antiphase boundary motion and its application to antiphase domain coarsening. Acta Metall 27(6):1085–109510.1016/0001-6160(79)90196-2 ↗
31. Zhang E, Yin M, Karniadakis GE (2020) Physics-informed neural networks for nonhomogeneous material identification in elasticity imaging, arXiv preprint arXiv:2009.04525
32. Abueidda DW, Lu Q, Koric S (2021) Meshless physics-informed deep learning method for three-dimensional solid mechanics. Int J Numer Meth Eng 122(23):7182–720110.1002/nme.6828 ↗
33. Kirby RC, Mitchell L (2019) Code generation for generally mapped finite elements. ACM Trans Math Softw (TOMS) 45(4):1–2310.1145/3361745 ↗
34. Vahab M, Haghighat E, Khaleghi M, Khalili N (2021) A physics informed neural network approach to solution and identification of biharmonic equations of elasticity, arXiv preprint arXiv:2108.0724310.1061/(ASCE)EM.1943-7889.0002062 ↗
35. Guo H, Zhuang X, Rabczuk T (2021) A deep collocation method for the bending analysis of kirchhoff plate, arXiv preprint arXiv:2102.02617
36. Zhu Y, Zabaras N, Koutsourelakis P-S, Perdikaris P (2019) Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data. J Comput Phys 394:56–8110.1016/j.jcp.2019.05.024 ↗
37. Kumar D, Koutsawa Y, Rauchs G, Marchi M, Kavka C, Belouettar S (2020) Efficient uncertainty quantification and management in the early stage design of composite applications. Compos Struct 251:112538. https://doi.org/10.1016/j.compstruct.2020.11253810.1016/j.compstruct.2020.112538 ↗
38. Butler R, Dodwell TJ, Haftka RT, Kim NH, Kim T, Kynaston S, Scheichl R (2015) Uncertainty quantification of composite structures with defects using multilevel monte carlo simulations. https://doi.org/10.2514/6.2015-159810.2514/6.2015-1598 ↗
39. Nabian MA, Meidani H (2018) A deep neural network surrogate for high-dimensional random partial differential equations, arXiv preprint arXiv:1806.0295710.1016/j.probengmech.2019.05.001 ↗
40. Kanoute P, Boso D, Chaboche J, Schrefler B (2009) Multiscale methods for composites: a review. Arch Comput Methods Eng 16(1):31–7510.1007/s11831-008-9028-8 ↗
41. Chiu P-H, Wong JC, Ooi C, Dao MH, Ong Y-S (2022) Can-pinn: a fast physics-informed neural network based on coupled-automatic-numerical differentiation method. Comput Methods Appl Mech Eng 395:114909
42. Kingma DP, Ba J (2014) Adam: a method for stochastic optimization, arXiv preprint arXiv:1412.6980
43. Liu DC, Nocedal J (1989) On the limited memory bfgs method for large scale optimization. Math Program 45(1):503–528
44. Halton JH (1960) On the efficiency of certain quasi-random sequences of points in evaluating multi-dimensional integrals. Numer Math 2(1):84–9010.1007/BF01386213 ↗
45. Nabian MA, Gladstone RJ, Meidani H (2021) Efficient training of physics-informed neural networks via importance sampling. Computer-Aided Civ Infrastruct Eng 36(8):962–97710.1111/mice.12685 ↗
46. Leiteritz R, Pflüger D (2021) How to avoid trivial solutions in physics-informed neural networks, arXiv preprint arXiv:2112.05620
47. Rahaman N, Baratin A, Arpit D, Draxler F, Lin M, Hamprecht F, Bengio Y, Courville A (2019) On the spectral bias of neural networks, In: International conference on machine learning, PMLR, pp 5301–5310
48. Huang G, Sun Y, Liu Z, Sedra D, Weinberger KQ (2016) Deep networks with stochastic depth. European conference on computer vision. Springer, Berlin, pp 646–661
49. Sitzmann V, Martel JN, Bergman AW, Lindell DB, Wetzstein G (2020) Implicit neural representations with periodic activation functions, arXiv preprint arXiv:2006.09661
50. Hennigh O, Narasimhan S, Nabian MA, Subramaniam A, Tangsali K, Fang Z, Rietmann M, Byeon W, Choudhry S (2021) Nvidia simnet$$^{{\rm TM}}$$: an ai-accelerated multi-physics simulation framework. International conference on computational science. Springer, Berlin, pp 447–461
51. Ramachandran P, Zoph B, Le QV (2017) Searching for activation functions, arXiv preprint arXiv:1710.05941
52. Klambauer G, Unterthiner T, Mayr A, Hochreiter S (2017) Self-normalizing neural networks, Adv Neural Inf Process Syst 30. https://doi.org/10.48550/arXiv.1706.0251510.48550/arXiv.1706.02515 ↗
53. Elfwing S, Uchibe E, Doya K (2018) Sigmoid-weighted linear units for neural network function approximation in reinforcement learning. Neural Netw 107:3–1110.1016/j.neunet.2017.12.012 ↗
54. Rao C, Sun H, Liu Y (2021) Physics-informed deep learning for computational elastodynamics without labeled data. J Eng Mech 147(8):0402104310.1061/(ASCE)EM.1943-7889.0001947 ↗
55. Falzone G, Falla GP, Wei Z, Zhao M, Kumar A, Bauchy M, Neithalath N, Pilon L, Sant G (2016) The influences of soft and stiff inclusions on the mechanical properties of cementitious composites. Cement Concr Compos 71:153–16510.1016/j.cemconcomp.2016.05.008 ↗
56. Peng T, Liu Y, Saxena A, Goebel K (2015) In-situ fatigue life prognosis for composite laminates based on stiffness degradation. Compos Struct 132:155–16510.1016/j.compstruct.2015.05.006 ↗
57. Fincato R, Tsutsumi S, Sakai T, Terada K (2019) Influence of hard and soft inclusions inside a ferritic matrix, In: MATEC web of conferences, vol 300, EDP Sciences, p 1000210.1051/matecconf/201930010002 ↗
58. Hughes TJ (2012) The finite element method: linear static and dynamic finite element analysis. Courier Corporation
59. Elguedj T, Bazilevs Y, Calo VM, Hughes TJ (2008) B and f projection methods for nearly incompressible linear and non-linear elasticity and plasticity using higher-order nurbs elements. Comput Methods Appl Mech Eng 197(33–40):2732–276210.1016/j.cma.2008.01.012 ↗
60. de Souza Neto E, Perić D, Dutko M, Owen D (1996) Design of simple low order finite elements for large strain analysis of nearly incompressible solids. Int J Solids Struct 33(20–22):3277–329610.1016/0020-7683(95)00259-6 ↗
61. Simo J, Taylor RL, Pister K (1985) Variational and projection methods for the volume constraint in finite deformation elasto-plasticity. Comput Methods Appl Mech Eng 51(1–3):177–20810.1016/0045-7825(85)90033-7 ↗
62. Jos KTG, Vinoy KJ (2018) Fem based methods for uncertainty quantification in electromagnetics. In: IEEE Indian conference on antennas and propogation (InCAP) pp 1–4. https://doi.org/10.1109/INCAP.2018.877075910.1109/INCAP.2018.8770759 ↗
63. Betz W, Papaioannou I, Straub D (2014) Numerical methods for the discretization of random fields by means of the karhunen-loève expansion. Comput Methods Appl Mech Eng 271:109–12910.1016/j.cma.2013.12.010 ↗
64. Fritzen F, Boehlke T (2011) Periodic three-dimensional mesh generation for particle reinforced composites with application to metal matrix composites. Int J Solids Struct 48(5):706–71810.1016/j.ijsolstr.2010.11.010 ↗
65. Molkov B, Pobedria B (1985) Effective characteristic of fibrous unidirectional composite with periodic structure. Mech Solids 2:119–129
66. Rodríguez-Ramos R, Berger H, Guinovart-Díaz R, López-Realpozo JC, Würkner M, Gabbert U, Bravo-Castillero J (2012) Two approaches for the evaluation of the effective properties of elastic composite with parallelogram periodic cells. Int J Eng Sci 58:2–1010.1016/j.ijengsci.2012.03.020 ↗
67. Watt JP, Davies GF, O’Connell RJ (1976) The elastic properties of composite materials. Rev Geophys 14(4):541–56310.1029/RG014i004p00541 ↗
68. Papathanasiou T, Ingber M, Mondy L, Graham A (1994) The effective elastic modulus of fiber-reinforced composites. J Compos Mater 28(4):288–30410.1177/002199839402800401 ↗
69. Xu Y, He Q-C, Gu S-T (2016) Effective elastic moduli of fiber-reinforced composites with interfacial displacement and stress jumps. Int J Solids Struct 80:146–15710.1016/j.ijsolstr.2015.10.031 ↗
70. Wong JC, Ooi C, Chiu P-H, Dao MH (2021) Improved surrogate modeling of fluid dynamics with physics-informed neural networks, arXiv preprint arXiv:2105.01838