Convolutional Neural Networks for the Localization of Plastic Velocity Gradient Tensor in Polycrystalline Microstructures
Source abstract
Abstract
Abstract Recent work has demonstrated the potential of convolutional neural networks (CNNs) in producing low-computational cost surrogate models for the localization of mechanical fields in two-phase microstructures. The extension of the same CNNs to polycrystalline microstructures is hindered by the lack of an efficient formalism for the representation of the crystal lattice orientation in the input channels of the CNNs. In this paper, we demonstrate the benefits of using generalized spherical harmonics (GSH) for addressing this challenge. A CNN model was successfully trained to predict the local plastic velocity gradient fields in polycrystalline microstructures subjected to a macroscopically imposed loading condition. Specifically, it is demonstrated that the proposed approach improves significantly the accuracy of the CNN models when compared with the direct use of Bunge–Euler angles to represent the crystal orientations in the input channels. Since the proposed approach implicitly satisfies the expected crystal symmetries in the specification of the input microstructure to the CNN, it opens new research directions for the adoption of CNNs in addressing a broad range of polycrystalline microstructure design and optimization problems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Non-Destructive Testing Techniques · Machine Learning in Materials Science · Model Reduction and Neural Networks
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332
School of Computational Science and Engineering, Georgia Institute of Technology, Atlanta, GA 30332
George W. Woodruff School of Mechanical Engineering, School of Computational Science and Engineering, Georgia Institute of Technology, Atlanta, GA 30332
References (71)
Show source-provided references
1. Overview of Constitutive Laws, Kinematics, Homogenization and Multiscale Methods in Crystal Plasticity Finite-Element Modeling: Theory, Experiments, Applications10.1016/j.actamat.2009.10.058 ↗
2. Xlvi. A Theory of the Plastic Distortion of a Polycrystalline Aggregate Under Combined Stresses10.1080/14786445108561065 ↗
3. Cxxviii. A Theoretical Derivation of the Plastic Properties of a Polycrystalline Face-Centred Metal10.1080/14786444108561385 ↗
4. On the Plastic Deformation of Polycrystals10.1016/0001-6160(61)90060-8 ↗
5. McDowell · 2009
6. An Inductive Design Exploration Method for Hierarchical Systems Design Under Uncertainty10.1080/03052150701742201 ↗
7. McDowell · 2008 · 20710.1007/978-1-4020-9741-6_14 ↗
8. Adams · 2012
9. On the Influence of State of Stress on Ductile Failure Initiation in High Strength Steels10.1016/0013-7944(77)90062-5 ↗
10. On the Ductile Enlargement of Voids in Triaxial Stress Fields10.1016/0022-5096(69)90033-7 ↗
11. Kalidindi · 2015
12. The Mesostructure-Properties Linkage in Polycrystals10.1016/S0079-6425(98)00002-4 ↗
13. A Second Gradient Theoretical Framework for Hierarchical Multiscale Modeling of Materials10.1016/j.ijplas.2010.05.006 ↗
14. Theory of Composites. Cambridge Monographs on Applied and Computational Mathematics10.1115/1.1553445 ↗
15. Second-Order Homogenization Estimates for Nonlinear Composites Incorporating Field Fluctuations: I—Theory10.1016/S0022-5096(01)00099-0 ↗
16. Microstructure-Sensitive Computational Modeling of Fatigue Crack Formation10.1016/j.ijfatigue.2010.01.003 ↗
17. Computational Damage Mechanics for Composite Materials Based on Mathematical Homogenization10.1002/(SICI)1097-0207(19990820)45:11<1657::AID-NME648>3.0.CO;2-H ↗
18. Qu · 200610.1002/9780470117835 ↗
19. Bounds for Effective Elastic Moduli of Disordered Materials10.1016/0022-5096(77)90009-6 ↗
20. Kröner · 1986 · 22910.1007/978-94-009-4181-6_8 ↗
21. A Computational Procedure for Rate-Independent Crystal Plasticity10.1016/0022-5096(96)00001-4 ↗
22. Crystallographic Texture Evolution in Bulk Deformation Processing of fcc Metals10.1016/0022-5096(92)80003-9 ↗
23. An Analysis of Nonuniform and Localized Deformation in Ductile Single Crystals10.1016/0001-6160(82)90005-0 ↗
24. Overview No. 42 Texture Development and Strain Hardening in Rate Dependent Polycrystals10.1016/0001-6160(85)90188-9 ↗
25. A Multiscale Model of Plasticity10.1016/S0749-6419(01)00044-4 ↗
26. A Predictive Discrete-Continuum Multiscale Model of Plasticity With Quantified Uncertainty10.1016/j.ijplas.2021.102935 ↗
27. Accelerating the Discovery of New dp Steel Using Machine Learning-Based Multiscale Materials Simulations10.1007/s11661-020-05764-7 ↗
28. Deep Learning10.1038/nature14539 ↗
29. Goodfellow · 2016
30. Representation and Calibration of Elastic Localization Kernels for a Broad Class of Cubic Polycrystals10.1016/j.actamat.2015.04.049 ↗
31. Calibrated Localization Relationships for Elastic Response of Polycrystalline Aggregates10.1016/j.actamat.2014.08.022 ↗
32. Reduced-Order Structure-Property Linkages for Polycrystalline Microstructures Based on 2-Point Statistics10.1016/j.actamat.2017.03.009 ↗
33. Prediction of Microscale Plastic Strain Rate Fields in Two-Phase Composites Subjected to an Arbitrary Macroscale Strain Rate Using the Materials Knowledge System Framework10.1016/j.actamat.2017.09.016 ↗
34. Localization Models for the Plastic Response of Polycrystalline Materials Using the Material Knowledge Systems Framework10.1088/1361-651X/ab37a5 ↗
35. Multi-Scale Modeling of Elastic Response of Three-Dimensional Voxel-Based Microstructure Datasets Using Novel DFT-Based Knowledge Systems10.1016/j.actamat.2010.01.007 ↗
36. Formulation and Calibration of Higher-Order Elastic Localization Relationships Using the MKS Approach10.1016/j.actamat.2011.04.005 ↗
37. Deep Learning Approaches for Mining Structure-Property Linkages in High Contrast Composites From Simulation Datasets10.1016/j.commatsci.2018.05.014 ↗
38. Establishing Structure-Property Localization Linkages for Elastic Deformation of Three-Dimensional High Contrast Composites Using Deep Learning Approaches10.1016/j.actamat.2018.12.045 ↗
39. Deep Learning Based Domain Knowledge Integration for Small Datasets: Illustrative Applications in Materials Informatics
40. Plastic Strain in Metals
41. Sachs · 1929 · 94
42. Bunge · 2013
43. Prediction of Cyclic Stress–Strain Property of Steels by Crystal Plasticity Simulations and Machine Learning10.3390/ma12223668 ↗
44. Deep Learning Based Predictive Modeling for Structure-Property Linkages10.1016/j.mtla.2019.100435 ↗
45. Application of Artificial Neural Networks in Micromechanics for Polycrystalline Metals10.1016/j.ijplas.2019.05.001 ↗
46. Procedures for Construction of Anisotropic Elastic–Plastic Property Closures for Face-Centered Cubic Polycrystals Using First-Order Bounding Relations10.1016/j.jmps.2006.01.010 ↗
47. Gel’fand · 2018
48. Orientation Imaging: The Emergence of a New Microscopy10.1007/BF02656503 ↗
49. Elastic–Plastic Property Closures for Hexagonal Close-Packed Polycrystalline Metals Using First-Order Bounding Theories10.1016/j.actamat.2006.12.010 ↗
50. Application of Microstructure Sensitive Design to Structural Components Produced From Hexagonal Polycrystalline Metals10.1016/j.commatsci.2007.12.002 ↗
51. Delineation of First-Order Closures for Plastic Properties Requiring Explicit Consideration of Strain Hardening and Crystallographic Texture Evolution10.1016/j.ijplas.2007.05.002 ↗
52. Segnet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation10.1109/TPAMI.2016.2644615 ↗
53. Image Segmentation-Based Multi-Focus Image Fusion Through Multi-Scale Convolutional Neural Network10.1109/ACCESS.2017.2735019 ↗
54. Convolutional Neural Network for Bio-Medical Image Segmentation With Hardware Acceleration10.1016/j.cogsys.2018.03.005 ↗
55. V-net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation
56. Using Convolutional Neural Networks to Identify Gravitational Lenses in Astronomical Images10.1093/mnras/stz1288 ↗
57. Extracting Grain Orientations From EBSD Patterns of Polycrystalline Materials Using Convolutional Neural Networks10.1017/S1431927618015131 ↗
58. Improving Direct Physical Properties Prediction of Heterogeneous Materials From Imaging Data Via Convolutional Neural Network and a Morphology-Aware Generative Model10.1016/j.commatsci.2018.03.074 ↗
59. Material Structure-Property Linkages Using Three-Dimensional Convolutional Neural Networks10.1016/j.actamat.2017.11.053 ↗
60. Improving Deep Neural Networks for Lvcsr Using Rectified Linear Units and Dropout
61. Multi-Column Deep Neural Networks for Image Classification
62. Toward Automatic Phenotyping of Developing Embryos From Videos10.1109/TIP.2005.852470 ↗
63. Learning Long-Range Vision for Autonomous Off-Road Driving10.1002/rob.20276 ↗
64. Natural Language Processing (Almost) From Scratch
65. Bottou · 2010 · 177
66. Dream. 3D: A Digital Representation Environment for the Analysis of Microstructure in 3D
67. Smith · 2009
68. Detailed Analyses of Grain–Scale Plastic Deformation in Columnar Polycrystalline Aluminium Using Orientation Image Mapping and Crystal Plasticity Models10.1098/rspa.2003.1260 ↗
69. A Survey of the Recent Architectures of Deep Convolutional Neural Networks10.1007/s10462-020-09825-6 ↗
70. Theory of the Backpropagation Neural Network
71. Adam: A Method for Stochastic Optimization