Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
126 tokens/sec
GPT-4o
47 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Guided Diffusion for Fast Inverse Design of Density-based Mechanical Metamaterials (2401.13570v2)

Published 24 Jan 2024 in cs.CE and cs.LG

Abstract: Mechanical metamaterial is a synthetic material that can possess extraordinary physical characteristics, such as abnormal elasticity, stiffness, and stability, by carefully designing its internal structure. To make metamaterials contain delicate local structures with unique mechanical properties, it is a potential method to represent them through high-resolution voxels. However, it brings a substantial computational burden. To this end, this paper proposes a fast inverse design method, whose core is an advanced deep generative AI algorithm, to generate voxel-based mechanical metamaterials. Specifically, we use the self-conditioned diffusion model, capable of generating a microstructure with a resolution of $1283$ to approach the specified homogenized tensor matrix in just 3 seconds. Accordingly, this rapid reverse design tool facilitates the exploration of extreme metamaterials, the sequence interpolation in metamaterials, and the generation of diverse microstructures for multi-scale design. This flexible and adaptive generative tool is of great value in structural engineering or other mechanical systems and can stimulate more subsequent research.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (46)
  1. Meza, L. R. et al. Resilient 3d hierarchical architected metamaterials. \JournalTitleProceedings of the National Academy of Sciences 112, 11502–11507 (2015).
  2. Highly-stretchable 3d-architected mechanical metamaterials. \JournalTitleScientific reports 6, 34147 (2016).
  3. Surjadi, J. U. et al. Mechanical metamaterials and their engineering applications. \JournalTitleAdvanced Engineering Materials 21, 1800864 (2019).
  4. Lattice metamaterials with mechanically tunable poisson’s ratio for vibration control. \JournalTitlePhysical Review Applied 7, 024012 (2017).
  5. Harnessing out-of-plane deformation to design 3d architected lattice metamaterials with tunable poisson’s ratio. \JournalTitleScientific Reports 7, 8949 (2017).
  6. A novel twofold symmetry architected metamaterials with high compressibility and negative poisson’s ratio. \JournalTitleAdvanced Engineering Materials 23, 2001041 (2021).
  7. Sequential metamaterials with alternating poisson’s ratios. \JournalTitleNature communications 13, 1041 (2022).
  8. Self-bridging metamaterials surpassing the theoretical limit of poisson’s ratios. \JournalTitleNature Communications 14, 4041 (2023).
  9. Geometric bounds for buckling-induced auxetic metamaterials undergoing large deformation. \JournalTitleApplied Mechanics and Materials 846, 547–552 (2016).
  10. Multi-stable mechanical metamaterials by elastic buckling instability. \JournalTitleJournal of materials science 54, 3509–3526 (2019).
  11. Yves, S. et al. Crystalline metamaterials for topological properties at subwavelength scales. \JournalTitleNature communications 8, 16023 (2017).
  12. Flexible mechanical metamaterials. \JournalTitleNature Reviews Materials 2, 1–11 (2017).
  13. Mechanical metamaterials associated with stiffness, rigidity and compressibility: A brief review. \JournalTitleProgress in Materials Science 94, 114–173 (2018).
  14. Mechanical metamaterials and beyond. \JournalTitleNature Communications 14, 6004 (2023).
  15. A review of homogenization and topology optimization i—homogenization theory for media with periodic structure. \JournalTitleComputers & Structures 69, 707–717 (1998).
  16. Topology optimization using petsc: An easy-to-use, fully parallel, open source topology optimization framework. \JournalTitleStructural and Multidisciplinary Optimization 51, 565–572 (2015).
  17. Large-scale topology optimization in 3d using parallel computing. \JournalTitleComputer methods in applied mechanics and engineering 190, 6201–6229 (2001).
  18. Giga-voxel computational morphogenesis for structural design. \JournalTitleNature 550, 84–86 (2017).
  19. High resolution topology optimization using graphics processing units (gpus). \JournalTitleStructural and Multidisciplinary Optimization 49, 315–325 (2014).
  20. Gpu parallel strategy for parameterized lsm-based topology optimization using isogeometric analysis. \JournalTitleStructural and Multidisciplinary Optimization 56, 413–434 (2017).
  21. A system for high-resolution topology optimization. \JournalTitleIEEE transactions on visualization and computer graphics 22, 1195–1208 (2015).
  22. An optimized, easy-to-use, open-source gpu solver for large-scale inverse homogenization problems. \JournalTitleStructural and Multidisciplinary Optimization 66, 207 (2023).
  23. Stainko, R. An adaptive multilevel approach to the minimal compliance problem in topology optimization. \JournalTitleCommunications in numerical methods in engineering 22, 109–118 (2006).
  24. Dynamic adaptive mesh refinement for topology optimization. \JournalTitlearXiv preprint arXiv:1009.4975 (2010).
  25. Connecting microstructures for multiscale topology optimization with connectivity index constraints. \JournalTitleJournal of Mechanical Design 140, 111417 (2018).
  26. Compatibility in microstructural optimization for additive manufacturing. \JournalTitleAdditive Manufacturing 26, 65–75 (2019).
  27. Data-driven design for metamaterials and multiscale systems: A review. \JournalTitleAdvanced Materials 2305254 (2023).
  28. A deep learning approach for inverse design of gradient mechanical metamaterials. \JournalTitleInternational Journal of Mechanical Sciences 240, 107920 (2023).
  29. Deep learning in mechanical metamaterials: From prediction and generation to inverse design. \JournalTitleAdvanced Materials 2302530 (2023).
  30. Controllable inverse design of auxetic metamaterials using deep learning. \JournalTitleMaterials & Design 211, 110178 (2021).
  31. Inverse-designed spinodoid metamaterials. \JournalTitlenpj Computational Materials 6, 73 (2020).
  32. Improving connectivity and accelerating multiscale topology optimization using deep neural network techniques. \JournalTitleStructural and Multidisciplinary Optimization 65, 126 (2022).
  33. Dl-msto+: A deep learning-based multi-scale topology optimization framework via positive definiteness ensured material representation network. \JournalTitleComputer Methods in Applied Mechanics and Engineering 415, 116276 (2023).
  34. Artificial intelligence-enabled smart mechanical metamaterials: advent and future trends. \JournalTitleInternational Materials Reviews 66, 365–393 (2021).
  35. Artificial intelligence in the design of innovative metamaterials: A comprehensive review. \JournalTitleInternational Journal of Precision Engineering and Manufacturing 1–20 (2023).
  36. Data-driven metamaterial design with laplace-beltrami spectrum as “shape-dna”. \JournalTitleStructural and multidisciplinary optimization 61, 2613–2628 (2020).
  37. Metaset: Exploring shape and property spaces for data-driven metamaterials design. \JournalTitleJournal of Mechanical Design 143, 031707 (2021).
  38. Lee, D. et al. t-METASET: Task-Aware Acquisition of Metamaterial Datasets Through Diversity-Based Active Learning. \JournalTitleJournal of Mechanical Design 145, 031704, 10.1115/1.4055925 (2022). https://asmedigitalcollection.asme.org/mechanicaldesign/article-pdf/145/3/031704/6938884/md_145_3_031704.pdf.
  39. Uncertainty quantification of microstructure variability and mechanical behavior of additively manufactured lattice structures. \JournalTitleComputer Methods in Applied Mechanics and Engineering 385, 114049 (2021).
  40. Inverting the structure–property map of truss metamaterials by deep learning. \JournalTitleProceedings of the National Academy of Sciences 119, e2111505119 (2022).
  41. Unifying the design space and optimizing linear and nonlinear truss metamaterials by generative modeling. \JournalTitleNature Communications 14, 7563 (2023).
  42. Best practices for convolutional neural networks applied to visual document analysis. In Icdar, vol. 3 (Edinburgh, 2003).
  43. Best practices for convolutional neural networks applied to visual document analysis. In Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings., 958–963, 10.1109/ICDAR.2003.1227801 (2003).
  44. Analog bits: Generating discrete data using diffusion models with self-conditioning. \JournalTitlearXiv preprint arXiv:2208.04202 (2022).
  45. Diffusion models beat gans on image synthesis. \JournalTitleAdvances in neural information processing systems 34, 8780–8794 (2021).
  46. Zheng, X.-Y. et al. Locally attentional sdf diffusion for controllable 3d shape generation. \JournalTitlearXiv e-prints arXiv–2305 (2023).
Citations (1)
View on Semantic Scholar

Summary

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

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

Tweets