Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
166 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

On the data-driven description of lattice materials mechanics (2310.20056v2)

Published 30 Oct 2023 in math.NA and cs.NA

Abstract: In the emerging field of mechanical metamaterials, using periodic lattice structures as a primary ingredient is relatively frequent. However, the choice of aperiodic lattices in these structures presents unique advantages regarding failure, e.g., buckling or fracture, because avoiding repeated patterns prevents global failures, with local failures occurring in turn that can beneficially delay structural collapse. Therefore, it is expedient to develop models for computing efficiently the effective mechanical properties in lattices from different general features while addressing the challenge of presenting topologies (or graphs) of different sizes. In this paper, we develop a deep learning model to predict energetically-equivalent mechanical properties of linear elastic lattices effectively. Considering the lattice as a graph and defining material and geometrical features on such, we show that Graph Neural Networks provide more accurate predictions than a dense, fully connected strategy, thanks to the geometrically induced bias through graph representation, closer to the underlying equilibrium laws from mechanics solved in the direct problem. Leveraging the efficient forward-evaluation of a vast number of lattices using this surrogate enables the inverse problem, i.e., to obtain a structure having prescribed specific behavior, which is ultimately suitable for multiscale structural optimization problems.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (116)
  1. Cellular solids: structure and properties. Cambridge University Press, 1999.
  2. Micro-architectured materials: past, present and future. In Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences, volume 466, pages 2495–2516, 2010.
  3. Ultralight, ultrastiff mechanical metamaterials. Science, 344:1373–1377, 2014.
  4. Nan Yang. Novel structural design method inspired by dna and origami. Results in Engineering, 4:100069, 2019.
  5. MF Ashby. Overview no. 92: Materials and shape. Acta metallurgica et materialia, 39(6):1025–1039, 1991.
  6. Materials selection in mechanical design. Le Journal de Physique IV, 3(C7):C7–1, 1993.
  7. A topology optimization method for design of negative permeability metamaterials. Structural and Multidisciplinary Optimization, 41:163–177, 2010.
  8. Wireless displacement sensing enabled by metamaterial probes for remote structural health monitoring. Sensors, 14(1):1691–1704, 2014.
  9. Homogenization methods of lattice materials. Encyclopedia, 2(2):1091–1102, 2022.
  10. Jörg Schröder. A numerical two-scale homogenization scheme: the fe2-method. In Plasticity and beyond: microstructures, crystal-plasticity and phase transitions, pages 1–64. Springer, 2014.
  11. A multiscale numerical homogenization-based method for the prediction of elastic properties of components produced with the fused deposition modelling process. Results in Engineering, 14:100409, 2022.
  12. A fast numerical method for computing the linear and nonlinear mechanical properties of composites. Comptes Rendus de l’Académie des sciences. Série II. Mécanique, physique, chimie, astronomie, 1994.
  13. Matti Schneider. A review of nonlinear fft-based computational homogenization methods. Acta Mechanica, 232(6):2051–2100, 2021.
  14. Challenges in two-scale computational homogenization of mechanical metamaterials. PAMM, 23(1):e202200139, 2023.
  15. Data-driven modeling and learning in science and engineering. Comptes Rendus Mécanique, 347(11):845–855, 2019.
  16. Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proceedings of the national academy of sciences, 113(15):3932–3937, 2016.
  17. Deep learning of thermodynamics-aware reduced-order models from data. Computer Methods in Applied Mechanics and Engineering, 379:113763, 2021.
  18. Ricardo Vinuesa. High-fidelity simulations in complex geometries: Towards better flow understanding and development of turbulence models. Results in Engineering, 11:100254, 2021.
  19. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational physics, 378:686–707, 2019.
  20. Training physics-informed neural networks: One learning to rule them all? Results in Engineering, 18:101023, 2023.
  21. Physics-informed graph neural galerkin networks: A unified framework for solving pde-governed forward and inverse problems. Computer Methods in Applied Mechanics and Engineering, 390:114502, 2022.
  22. Physics informed and data-based augmented learning in structural health diagnosis. Computer Methods in Applied Mechanics and Engineering, 414:116186, 2023.
  23. Ill-posed problems of mathematical physics and analysis, volume 64. American Mathematical Soc., 1986.
  24. Statistical and computational inverse problems, volume 160. Springer Science & Business Media, 2006.
  25. Deep learning for topology optimization of 2d metamaterials. Materials & Design, 196:109098, 2020.
  26. A gentle introduction to graph neural networks. Distill, 6(9):e33, 2021.
  27. A graph convolutional autoencoder approach to model order reduction for parametrized pdes. Journal of Computational Physics, page 112762, 2024.
  28. A comprehensive survey of graph neural networks for knowledge graphs. IEEE Access, 10:75729–75741, 2022.
  29. Graph neural networks for social recommendation. In The World Wide Web Conference, WWW ’19. ACM, May 2019.
  30. Neural message passing for quantum chemistry, 2017.
  31. A graph neural network framework for causal inference in brain networks. Scientific Reports, 11(1), April 2021.
  32. On the use of mechanics-informed models to structural engineering systems: Application of graph neural networks for structural analysis. Structures, 59:105712, January 2024.
  33. A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems, 32(1):4–24, January 2021.
  34. Graph neural networks: A review of methods and applications. AI Open, 1:57–81, 2020.
  35. Grand: Graph neural diffusion, 2021.
  36. S Choukir and CV Singh. Role of topology in dictating the fracture toughness of mechanical metamaterials. International Journal of Mechanical Sciences, 241:107945, 2023.
  37. Enhanced vibration suppression using diatomic acoustic metamaterial with negative stiffness mechanism. Engineering Structures, 271:114939, 2022.
  38. Aperiodic-metamaterial-based absorber. APL Materials, 5(9), September 2017.
  39. A design strategy to match the band gap of periodic and aperiodic metamaterials. Scientific Reports, 10(1), October 2020.
  40. Inverting the structure–property map of truss metamaterials by deep learning. Proceedings of the National Academy of Sciences, 119(1), December 2021.
  41. Periodic truss structures. Journal of the Mechanics and Physics of Solids, 96:184–203, November 2016.
  42. Mark C. Messner. Optimal lattice-structured materials. Journal of the Mechanics and Physics of Solids, 96:162–183, November 2016.
  43. Reexamining the mechanical property space of three-dimensional lattice architectures. Acta Materialia, 140:424–432, November 2017.
  44. Effective elastic properties and initial yield surfaces of two 3d lattice structures. International Journal of Mechanical Sciences, 138-139:146–158, April 2018.
  45. Deep learning-assisted elastic isotropy identification for architected materials. Extreme Mechanics Letters, 43:101173, February 2021.
  46. S. Torquato. Statistical description of microstructures. Annual Review of Materials Research, 32(1):77–111, August 2002.
  47. The correlation between statistical descriptors of heterogeneous materials. Computer Methods in Applied Mechanics and Engineering, 384:113948, October 2021.
  48. Elizabeth Munch. A user’s guide to topological data analysis. Journal of Learning Analytics, 4(2), July 2017.
  49. An introduction to topological data analysis: Fundamental and practical aspects for data scientists. Frontiers in Artificial Intelligence, 4, September 2021.
  50. Topological data analysis of contagion maps for examining spreading processes on networks. Nature Communications, 6(1), July 2015.
  51. Describing and modeling rough composites surfaces by using topological data analysis and fractional brownian motion. Polymers, 15(6):1449, March 2023.
  52. Non linear constitutive models for lattice materials. Journal of the Mechanics and Physics of Solids, 64:44–60, March 2014.
  53. Non-affinity in multi-material mechanical metamaterials. Scientific Reports, 10(1), July 2020.
  54. Learning the nonlinear dynamics of mechanical metamaterials with graph networks. International Journal of Mechanical Sciences, 238:107835, January 2023.
  55. B. Hassani and E. Hinton. A review of homogenization and topology optimization i—homogenization theory for media with periodic structure. Computers & Structures, 69(6):707–717, December 1998.
  56. B. Hassani and E. Hinton. A review of homogenization and topology opimization II—analytical and numerical solution of homogenization equations. Computers & Structures, 69(6):719–738, December 1998.
  57. B. Hassani and E. Hinton. A review of homogenization and topology optimization III—topology optimization using optimality criteria. Computers & Structures, 69(6):739–756, December 1998.
  58. Asymptotic homogenization models for pantographic lattices with variable order rotational resistance at pivots. Journal of the Mechanics and Physics of Solids, 134:103718, January 2020.
  59. Homogenization of discrete thin structures. Nonlinear Analysis, 231:112951, June 2023.
  60. Topology optimization and de-homogenization of graded lattice structures based on asymptotic homogenization. Composite Structures, 277:114633, December 2021.
  61. Lattice topology homogenization and crack propagation through finite element analyses. Procedia Structural Integrity, 28:637–647, 2020.
  62. Spectral graph theory for characterization and homogenization of grain boundary networks. Acta Materialia, 146:42–54, March 2018.
  63. On dynamic random graphs with degree homogenization via anti-preferential attachment probabilities. Physica D: Nonlinear Phenomena, 414:132689, December 2020.
  64. A meta-path graph-based graph homogenization framework for machine fault diagnosis. Engineering Applications of Artificial Intelligence, 121:105960, May 2023.
  65. Dominik Dold and Derek Aranguren van Egmond. Differentiable graph-structured models for inverse design of lattice materials. Cell Reports Physical Science, page 101586, September 2023.
  66. AI-aided design of multiscale lattice metastructures for controllable anisotropy. Materials & Design, 223:111254, November 2022.
  67. Deep learning model to predict fracture mechanisms of graphene. npj 2D Materials and Applications, 5(1), April 2021.
  68. The graph neural network model. IEEE Transactions on Neural Networks, 20(1):61–80, January 2009.
  69. Graph neural networks and implicit neural representation for near-optimal topology prediction over irregular design domains. Engineering Applications of Artificial Intelligence, 123:106284, August 2023.
  70. Graph neural networks (GNNs) based accelerated numerical simulation. Engineering Applications of Artificial Intelligence, 123:106370, August 2023.
  71. Learning mechanically driven emergent behavior with message passing neural networks. Computers & Structures, 270:106825, October 2022.
  72. Prediction and control of fracture paths in disordered architected materials using graph neural networks. Communications Engineering, 2(1), June 2023.
  73. Dynamic and adaptive mesh-based graph neural network framework for simulating displacement and crack fields in phase field models. Mechanics of Materials, 186:104789, November 2023.
  74. Materials fatigue prediction using graph neural networks on microstructure representations. Scientific Reports, 13(1), August 2023.
  75. Predicting stress, strain and deformation fields in materials and structures with graph neural networks. Scientific Reports, 12(1), December 2022.
  76. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational Physics, 378:686–707, February 2019.
  77. Physics-informed machine learning. Nature Reviews Physics, 3(6):422–440, May 2021.
  78. GINNs: Graph-informed neural networks for multiscale physics. Journal of Computational Physics, 433:110192, May 2021.
  79. Physics-informed graph convolutional neural network for modeling geometry-adaptive steady-state natural convection. International Journal of Heat and Mass Transfer, 216:124593, December 2023.
  80. Physics-informed graph neural networks accelerating microneedle simulations towards novelty of micro-nano scale materials discovery. Engineering Applications of Artificial Intelligence, 126:106894, November 2023.
  81. Physics-informed graph neural network emulation of soft-tissue mechanics. Computer Methods in Applied Mechanics and Engineering, 417:116351, December 2023.
  82. Physics-informed graph neural network for spatial-temporal production forecasting. Geoenergy Science and Engineering, 223:211486, April 2023.
  83. Non-deterministic methods and surrogates in the design of rockfill dams. Applied Sciences, 11(8):3699, April 2021.
  84. Hannah Jordan. A Non-Deterministic Deep Learning Based Surrogate for Ice Sheet Modelling. Master of Science (MS) Thesis, Department of Computer Science, University of Montana, 2022.
  85. A spectral surrogate model for stochastic simulators computed from trajectory samples. Computer Methods in Applied Mechanics and Engineering, 406:115875, March 2023.
  86. Past, current and future trends and challenges in non-deterministic fracture mechanics: A review. Computer Methods in Applied Mechanics and Engineering, 412:116102, July 2023.
  87. Generalized uncertainty in surrogate models for concrete strength prediction. Engineering Applications of Artificial Intelligence, 122:106155, June 2023.
  88. Neural network-assisted probabilistic creep-fatigue assessment of hydrogenation reactor with physics-based surrogate model. International Journal of Pressure Vessels and Piping, 206:105051, December 2023.
  89. Surrogate modeling and global sensitivity analysis for the simultaneous growth of multiple hydraulic fractures. Computers and Geotechnics, 162:105709, October 2023.
  90. Physics-based surrogate model for reinforced concrete corrosion simulation. Results in Engineering, 16:100659, 2022.
  91. A. Bowyer. Computing dirichlet tessellations. The Computer Journal, 24(2):162–166, February 1981.
  92. D. F. Watson. Computing the n-dimensional delaunay tessellation with application to voronoi polytopes. The Computer Journal, 24(2):167–172, February 1981.
  93. Deep learning framework for material design space exploration using active transfer learning and data augmentation. npj Computational Materials, 7(1), September 2021.
  94. Sabrina Chin-yun Shen and Markus J. Buehler. Nature-inspired architected materials using unsupervised deep learning. Communications Engineering, 1(1), November 2022.
  95. Gam: General auxetic metamaterial with tunable 3d auxetic behavior using the same unit cell boundary connectivity. Materials, 16(9):3473, 2023.
  96. Designing architected materials for mechanical compression via simulation, deep learning, and experimentation. npj Computational Materials, 9(1), May 2023.
  97. Obtaining auxetic and isotropic metamaterials in counterintuitive design spaces: an automated optimization approach and experimental characterization. npj Computational Materials, 10(1), January 2024.
  98. An artificial neural network framework for reduced order modeling of transient flows. Communications in Nonlinear Science and Numerical Simulation, 77:271–287, 2019.
  99. Pattern recognition and machine learning, volume 4. Springer, 2006.
  100. Convex optimization. Cambridge university press, 2004.
  101. Topology optimization approach for functionally graded metamaterial components based on homogenization of mechanical variables. Computers & Structures, 289:107151, 2023.
  102. Auxetic metamaterials from disordered networks. Proceedings of the National Academy of Sciences, 115(7), January 2018.
  103. Bioinspired structural hydrogels with highly ordered hierarchical orientations by flow-induced alignment of nanofibrils. Nature Communications, 15(1), January 2024.
  104. Synthesizing controlled microstructures of porous media using generative adversarial networks and reinforcement learning. Scientific Reports, 12(1), May 2022.
  105. Design method of lightweight metamaterials with arbitrary poisson’s ratio. Materials, 11(9):1574, September 2018.
  106. Minimal-surface-based multiphase metamaterials with highly variable stiffness. Materials and Design, 237:112548, January 2024.
  107. Controllable inverse design of auxetic metamaterials using deep learning. Materials and Design, 211:110178, December 2021.
  108. Unifying the design space and optimizing linear and nonlinear truss metamaterials by generative modeling. Nature Communications, 14(1), November 2023.
  109. Janusz S Przemieniecki. Theory of matrix structural analysis. Courier Corporation, 1985.
  110. Multilayer feedforward networks are universal approximators. Neural networks, 2(5):359–366, 1989.
  111. Self-normalizing neural networks. Advances in neural information processing systems, 30, 2017.
  112. Timothy Dozat. Incorporating nesterov momentum into adam. 2016.
  113. Automatic differentiation in machine learning: a survey. Journal of Marchine Learning Research, 18:1–43, 2018.
  114. Deep-learning-based inverse design of three-dimensional architected cellular materials with the target porosity and stiffness using voxelized voronoi lattices. Science and Technology of Advanced Materials, 24(1), January 2023.
  115. Inverse machine learning framework for optimizing lightweight metamaterials. Materials & Design, 208:109937, October 2021.
  116. The updated properties model (upm): A topology optimization algorithm for the creation of macro–micro optimized structures with variable stiffness. Finite Elements in Analysis and Design, 223:103970, 2023.
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now