Feed-Forward Neural Networks for Failure Mechanics Problems
Source abstract
Abstract
This work addresses an efficient neural network (NN) representation for the phase-field modeling of isotropic brittle fracture. In recent years, data-driven approaches, such as neural networks, have become an active research field in mechanics. In this contribution, deep neural networks—in particular, the feed-forward neural network (FFNN)—are utilized directly for the development of the failure model. The verification and generalization of the trained models for elasticity as well as fracture behavior are investigated by several representative numerical examples under different loading conditions. As an outcome, promising results close to the exact solutions are produced.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Numerical methods in engineering · Fatigue and fracture mechanics · Non-Destructive Testing Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Institute of Continuum Mechanics, Leibniz University Hannover, 30823 Garbsen, Germany
Funding
Deutsche Forschungsgemeinschaft
References (79)
Show source-provided references
1. Wriggers, P. (2008). Nonlinear Finite Elements, Springer.
2. Isogeometric analysis: CAD, finite elements, NURBS, exact geometry and mesh refinement10.1016/j.cma.2004.10.008 ↗
3. Basic principles of virtual element methods10.1142/S0218202512500492 ↗
4. A data-driven approach to nonlinear elasticity10.1016/j.compstruc.2017.07.031 ↗
5. Data-based derivation of material response10.1016/j.cma.2017.11.013 ↗
6. Model-free data-driven inelasticity10.1016/j.cma.2019.02.016 ↗
7. Data-driven fracture mechanics10.1016/j.cma.2020.113390 ↗
8. Bahmani, B., and Sun, W. (2020). An accelerated hybrid data-driven/model-based approach for poroelasticity problems with multi-fidelity multi-physics data. arXiv.10.1016/j.cma.2021.113868 ↗
9. Eggersmann, R., Stainier, L., Ortiz, M., and Reese, S. (2020). Efficient Data Structures for Model-free Data-Driven Computational Mechanics. arXiv.10.1016/j.cma.2021.113855 ↗
10. Bishop, C. (2016). Pattern Recognition and Machine Learning, Springer. Information Science and Statistics.
11. Fuhg, J.N., Böhm, C., Bouklas, N., Fau, A., Wriggers, P., and Marino, M. (2021). Model-data-driven constitutive responses: Application to a multiscale computational framework. arXiv.10.1016/j.ijengsci.2021.103522 ↗
12. Self-learning finite elements for inverse estimation of thermal constitutive models10.1016/j.ijheatmasstransfer.2006.01.031 ↗
13. Artificial neural network as an incremental non-linear constitutive model for a finite element code10.1016/S0045-7825(03)00350-5 ↗
14. Artificial Neural Networks in numerical modelling of composites10.1016/j.cma.2008.12.036 ↗
15. Neural network constitutive modelling for non-linear characterization of anisotropic materials10.1002/nme.2999 ↗
16. Palau, T., Kuhn, A., Nogales, S., Böhm, H., and Rauh, A. (2012). A neural network based elasto-plasticity material model. ECCOMAS 2012—European Congress on Computational Methods in Applied Sciences and Engineering, ECCOMAS Proceeding TU. e-Book Full Papers.
17. Incorporating neural network material models within finite element analysis for rheological behavior prediction10.1115/1.2389004 ↗
18. Machine learning materials physics: Multi-resolution neural networks learn the free energy and nonlinear elastic response of evolving microstructures10.1016/j.cma.2020.113362 ↗
19. DeepXDE: A deep learning library for solving differential equations10.1137/19M1274067 ↗
20. Knowledge-based modeling of material behavior with neural networks10.1061/(ASCE)0733-9399(1991)117:1(132) ↗
21. On self-learning finite element codes based on monitored response of structures10.1016/S0266-352X(00)00016-1 ↗
22. Neural networks in mechanics of structures and materials—New results and prospects of applications10.1016/S0045-7949(01)00083-9 ↗
23. Waszczyszyn, Z. (2000, January 11–14). Neural Networks in Plasticity—Some New Results and Prospects of Applications. Proceedings of the ECCOMAS—European Congress on Computational Methods in Applied Sciences and Engineering, Barcelona, Spain.
24. Neural networks in computational mechanics10.1007/BF02818935 ↗
25. Generalised hardening plasticity approximated via anisotropic elasticity: A neural network approach10.1016/0045-7825(94)00769-J ↗
26. Neural networks for computing in fracture mechanics. Methods and prospects of applications10.1016/0045-7825(93)90191-Y ↗
27. Göküzüm, F.S., Nguyen, L.T.K., and Keip, M.A. (2019). An Artificial Neural Network Based Solution Scheme for Periodic Computational Homogenization of Electrostatic Problems. Math. Comput. Appl., 24.10.3390/mca24020040 ↗
28. Fast simulations for solving fracture mechanics inverse problems using POD-RBF XIGA and Jaya algorithm10.1016/j.engfracmech.2018.09.032 ↗
29. A deep energy method for finite deformation hyperelasticity10.1016/j.euromechsol.2019.103874 ↗
30. The neural particle method—An updated Lagrangian physics informed neural network for computational fluid dynamics10.1016/j.cma.2020.113127 ↗
31. A hybrid approach to simulate the homogenized irreversible elastic–plastic deformations and damage of foams by neural networks10.1016/j.ijplas.2019.11.003 ↗
32. A surrogate model for computational homogenization of elastostatics at finite strain using high-dimensional model representation-based neural network10.1002/nme.6493 ↗
33. Data Mining and Machine Learning Methods Applied to 3 A Numerical Clinching Model
34. An intelligent nonlinear meta element for elastoplastic continua: Deep learning using a new Time-distributed Residual U-Net architecture10.1016/j.cma.2020.113088 ↗
35. Application of artificial neural networks for the prediction of interface mechanics: A study on grain boundary constitutive behavior10.1186/s40323-019-0138-7 ↗
36. An efficient optimization approach for designing machine learning models based on genetic algorithm10.1007/s00521-020-05035-x ↗
37. SO(3)-invariance of informed-graph-based deep neural network for anisotropic elastoplastic materials10.1016/j.cma.2020.112875 ↗
38. Improved ANN technique combined with Jaya algorithm for crack identification in plates using XIGA and experimental analysis10.1016/j.tafmec.2020.102554 ↗
39. An energy approach to the solution of partial differential equations in computational mechanics via machine learning: Concepts, implementation and applications10.1016/j.cma.2019.112790 ↗
40. DNN2: A hyper-parameter reinforcement learning game for self-design of neural network based elasto-plastic constitutive descriptions10.1016/j.compstruc.2021.106505 ↗
41. An offline multi-scale unsaturated poromechanics model enabled by self-designed/self-improved neural networks10.1002/nag.3196 ↗
42. Zhang, X., and Garikipati, K. (2021). Bayesian neural networks for weak solution of PDEs with uncertainty quantification. arXiv.
43. Crack prediction in pipeline using ANN-PSO based on numerical and experimental modal analysis
44. Autoprogressive training of neural network constitutive models10.1002/(SICI)1097-0207(19980515)42:1<105::AID-NME356>3.0.CO;2-V ↗
45. Neural Network Material Modelling10.1080/13921525.1999.10531472 ↗
46. Numerical implementation of a neural network based material model in finite element analysis10.1002/nme.905 ↗
47. A machine learning based plasticity model using proper orthogonal decomposition10.1016/j.cma.2020.113008 ↗
48. Prediction and optimization of mechanical properties of composites using convolutional neural networks10.1016/j.compstruct.2019.111264 ↗
49. Geometric deep learning for computational mechanics Part I: Anisotropic Hyperelasticity10.1016/j.cma.2020.113299 ↗
50. Deep learning approaches for mining structure-property linkages in high contrast composites from simulation datasets10.1016/j.commatsci.2018.05.014 ↗
51. Predicting the mechanical response of oligocrystals with deep learning10.1016/j.commatsci.2019.109099 ↗
52. Three-dimensional convolutional neural network (3D-CNN) for heterogeneous material homogenization10.1016/j.commatsci.2020.109850 ↗
53. Goodfellow, I., Bengio, Y., Courville, A., and Bengio, Y. (2016). Deep Learning, MIT Press.
54. Training Feedforward Networks with the Marquardt Algorithm10.1109/72.329697 ↗
55. Korelc, J., and Wriggers, P. (2016). Automation of Finite Element Methods, Springer.10.1007/978-3-319-39005-5 ↗
56. Thermodynamically consistent phase-field models of fracture: Variational principles and multi-field FE implementations10.1002/nme.2861 ↗
57. Towards Phase Field Modeling of Ductile Fracture in Gradient-Extended Elastic-Plastic Solids10.1002/pamm.201410193 ↗
58. A review on phase-field models of brittle fracture and a new fast hybrid formulation10.1007/s00466-014-1109-y ↗
59. A continuum phase field model for fracture10.1016/j.engfracmech.2010.08.009 ↗
60. Thermodynamically consistent algorithms for a finite-deformation phase-field approach to fracture10.1002/nme.4709 ↗
61. Aldakheel, F. (2016). Mechanics of Nonlocal Dissipative Solids: Gradient Plasticity and Phase Field Modeling of Ductile Fracture. [Ph.D. Thesis, University of Stuttgart].
62. A phase-field approach to model fracture of arterial walls: Theory and finite element analysis10.1016/j.cma.2016.04.007 ↗
63. Revisiting the problem of a crack impinging on an interface: A modeling framework for the interaction between the phase field approach for brittle fracture and the interface cohesive zone model10.1016/j.cma.2017.04.004 ↗
64. A higher-order phase-field model for brittle fracture: Formulation and analysis within the isogeometric analysis framework10.1016/j.cma.2014.01.016 ↗
65. Phase field modeling of fracture in anisotropic brittle solids10.1016/j.ijnonlinmec.2017.06.018 ↗
66. A phase-field approach embedded in the Theory of Porous Media for the description of dynamic hydraulic fracturing10.1016/j.cma.2016.10.045 ↗
67. Numerical evaluation of the phase-field model for brittle fracture with emphasis on the length scale10.1007/s00466-017-1373-8 ↗
68. Wick, T. (2020). Multiphysics Phase-Field Fracture: Modeling, Adaptive Discretizations, and Solvers, Walter de Gruyter GmbH & Co. KG.10.1515/9783110497397 ↗
69. Elasto-plastic phase-field model of hydraulic fracture in saturated binary porous media10.1615/IntJMultCompEng.2019027066 ↗
70. A residual control staggered solution scheme for the phase-field modeling of brittle fracture10.1016/j.engfracmech.2018.09.027 ↗
71. A finite-strain phase-field approach to ductile failure of frictional materials10.1016/j.ijsolstr.2019.02.006 ↗
72. A phase-field fracture model based on strain gradient elasticity10.1016/j.engfracmech.2019.106648 ↗
73. Formulation and implementation of strain rate-dependent fracture toughness in context of the phase-field method10.1002/nme.6207 ↗
74. Phase-field models for fatigue crack growth10.1016/j.tafmec.2019.102282 ↗
75. Phase-field modeling of porous-ductile fracture in non-linear thermo-elasto-plastic solids10.1016/j.cma.2019.112730 ↗
76. A phase-field model for fracture of unidirectional fiber-reinforced polymer matrix composites10.1007/s00466-019-01812-1 ↗
77. A phase field framework for capillary-induced fracture in unsaturated porous media: Drying-induced vs. hydraulic cracking10.1016/j.cma.2019.112647 ↗
78. Seiler, M., Keller, S., Kashaev, N., Klusemann, B., and Kästner, M. (2020). Phase-field modelling for fatigue crack growth under laser-shock-peening-induced residual stresses. Arch. Appl. Mech.10.1007/s00419-021-01897-2 ↗
79. Phase Field Modeling of Fracture in Multi-Physics Problems. Part II. Brittle-to-Ductile Failure Mode Transition and Crack Propagation in Thermo-Elastic-Plastic Solids10.1016/j.cma.2014.11.017 ↗