De novo composite design based on machine learning algorithm
Source abstract
Abstract
Abstract Composites are widely used to create tunable materials to achieve superior mechanical properties. Brittle materials fail catastrophically in the presence of cracks. Incorporating softer constituents into brittle materials can alleviate stress concentration, leading to tougher and stronger composites. However, searching for the optimal designs of composites is extremely challenging due to the astronomical number of possible material and geometry combinations. Here, we apply machine learning to a composite system and demonstrate its capacity to accurately and efficiently predict mechanical properties including toughness and strength. The method we used incorporates machine learning techniques to generate optimal designs with orders of magnitude better than the mean properties of the input training data, and at a much lower computational cost compared to exhaustive methods. Additionally, the method can further rebuild the detailed performances of the designs, even when this information is lost in the training process. The results demonstrate the ability of machine learning to search for optimal designs with very limited training data. To demonstrate the application of machine learning to composite design, we optimize a large-scale system not tractable by an exhaustive brute force approach and show that it is a promising tool towards composite design. This work offers a new perspective in the exploration of design spaces and accelerating the discovery of new functional, customizable composites.
Source: openaire · Quality: candidate unverified · Source record ↗
Research context
OpenAlex topics: Machine Learning in Materials Science · Additive Manufacturing and 3D Printing Technologies · Manufacturing Process and Optimization
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (43)
Show source-provided references
1. Biological and bioinspired composites with spatially tunable heterogeneous architectures10.1002/adfm.201300340 ↗
2. Chawla · 2012
3. Nacre-inspired design of graphene oxide–polydopamine nanocomposites for enhanced mechanical properties and multi-functionalities10.1088/2399-1984/aa6aed ↗
4. S.W. Tsai, Composites design, in: Think Composites, P. O. Box 581, Dayton, Ohio 45419, USA, 1988. p. 583.
5. Ceramic matrix composites
6. Chung · 2012
7. Mallick · 2007
8. Gibson · 2010
9. Hierarchically enhanced impact resistance of bioinspired composites10.1002/adma.201700060 ↗
10. Bone-inspired materials by design: toughness amplification observed using 3D printing and testing10.1002/adem.201600143 ↗
11. 3D printed bio-inspired prototypes and analytical models for structured suture interfaces with geometrically-tuned deformation and failure behavior10.1016/j.jmps.2014.08.011 ↗
12. MetaMesh: A hierarchical computational model for design and fabrication of biomimetic armored surfaces10.1016/j.cad.2014.05.005 ↗
13. Printing nature: Unraveling the role of nacre’s mineral bridges10.1016/j.jmbbm.2017.05.007 ↗
14. Three-dimensional-printing of bio-inspired composites10.1115/1.4032423 ↗
15. Biomimetic additive manufactured polymer composites for improved impact resistance10.1016/j.eml.2016.09.006 ↗
16. 3D-printing of lightweight cellular composites10.1002/adma.201401804 ↗
17. Biocompatibility and osteogenesis of biomimetic nano-hydroxyapatite/polyamide composite scaffolds for bone tissue engineering10.1016/j.biomaterials.2007.04.014 ↗
18. Additive manufacturing of tissues and organs10.1016/j.progpolymsci.2011.11.007 ↗
19. Genetic algorithms in structural topology optimization
20. Bendsoe · 2013
21. X. Guo, H. Gao, Bio-inspired material design and optimization, in: Proc. IUTAM Symposium on Topological Design Optimization of Structures, Machines and Materials, Springer, pp. 439–453.10.1007/1-4020-4752-5_43 ↗
22. Material optimization: bridging the gap between conceptual and preliminary design10.1016/S1270-9638(01)01125-7 ↗
23. Multiple-material topology optimization of compliant mechanisms created via polyjet three-dimensional printing10.1115/1.4028439 ↗
24. Numerical instabilities in topology optimization: a survey on procedures dealing with checkerboards, mesh-dependencies and local minima10.1007/BF01214002 ↗
25. Optimization of composite fracture properties: Method, validation, and applications10.1115/1.4033381 ↗
26. Mastering the game of go with deep neural networks and tree search10.1038/nature16961 ↗
27. J. Tromp, Number of legal Go positions.
28. Michalski · 2013
29. Murphy · 2012
30. Selection of relevant features and examples in machine learning10.1016/S0004-3702(97)00063-5 ↗
31. Kononenko · 2007
32. Witten · 2016
33. Face recognition by independent component analysis10.1109/TNN.2002.804287 ↗
34. A review of machine learning approaches to spam filtering10.1016/j.eswa.2009.02.037 ↗
35. A framework for data-driven analysis of materials under uncertainty: Countering the curse of dimensionality10.1016/j.cma.2017.03.037 ↗
36. Post-buckling optimisation of composite stiffened panels using neural networks10.1016/S0263-8223(02)00053-3 ↗
37. The reduced model multiscale method (R3M) for the non-linear homogenization of hyperelastic media at finite strains10.1016/j.jcp.2006.09.019 ↗
38. Machine learning strategy for accelerated design of polymer dielectrics10.1038/srep20952 ↗
39. Accelerating materials property predictions using machine learning10.1038/srep02810 ↗
40. Machine-learning-based mechanical properties prediction in foundry production
41. M. Abadi, A. Agarwal, P. Barham, E. Brevdo, Z. Chen, C. Citro, G.S. Corrado, A. Davis, J. Dean, M. Devin, Tensorflow: Large-scale machine learning on heterogeneous distributed systems, 2016. arXiv preprint arXiv:1603.04467.
42. D. Brackett, I. Ashcroft, R. Hague, Topology optimization for additive manufacturing, in: Proc. Proceedings of the Solid Freeform Fabrication Symposium, Austin, TX, S. pp. 348–362.
43. Algorithm-driven design of fracture resistant composite materials realized through additive manufacturing10.1016/j.addma.2017.07.002 ↗