Implicit geometric descriptor-enabled ANN Framework for a unified structure-property relationship in architected nanofibrous materials
Source abstract
Abstract
Hierarchically architected nanofibrous materials, such as the vertically aligned carbon nanotube (VACNT) foams, draw their exceptional mechanical properties from the interplay of nanoscale size effects and inter-nanotube interactions within and across architectures. However, the distinct effects of these mechanisms, amplified by the architecture, on different mechanical properties remain elusive, limiting their independent tunability for targeted property combinations. Reliance on architecture-specific explicit design parameters further inhibits the development of a unified structure-property relationship rooted in those nanoscale mechanisms. Here, we introduce two implicit geometric descriptors -- multi-component shape invariants (MCSI) -- in an artificial neural network (ANN) framework to establish a unified structure-property relationship that governs diverse architectures. The MCSIs effectively capture the key nanoscale mechanisms that give rise to the bulk mechanical properties such as specific-energy absorption, peak stress, and average modulus. Exploiting their ability to predict mechanical properties for designs that are even outside of the training data, we propose generalized design strategies to achieve desired mechanical property combinations in architected VACNT foams. Such implicit descriptor-enabled ANN frameworks can guide the accelerated and tractable design of complex hierarchical materials for applications ranging from shock-absorbing layers in extreme environments to functional components in soft robotics.
Source: openaire · Quality: candidate unverified · Source record ↗
Research context
OpenAlex topics: Calcium Carbonate Crystallization and Inhibition · Material Selection and Properties · Building materials and conservation
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (48)
Show source-provided references
1. The conflicts between strength and toughness10.1038/nmat3115 ↗
2. Bioinspired structural materials10.1038/nmat4089 ↗
3. Nanoscale deformation mechanisms in bone10.1021/nl051584b ↗
4. Nature’s hierarchical materials10.1016/j.pmatsci.2007.06.001 ↗
5. Strong and tough fibrous hydrogels reinforced by multiscale hierarchical structures with multimechanisms10.1126/sciadv.adf7075 ↗
6. Biomimetic cellular metals—using hierarchical structuring for energy absorption10.1088/1748-3190/11/4/045002 ↗
7. Analysis of uniaxial compression of vertically aligned carbon nanotubes10.1016/j.jmps.2011.05.002 ↗
8. Shock formation and rate effects in impacted carbon nanotube foams10.1016/j.carbon.2014.12.006 ↗
9. Carbon nanotubes with temperature-invariant viscoelasticity from–196 to 1000 c10.1126/science.1194865 ↗
10. Modeling frequency-and temperature-invariant dissipative behaviors of randomly entangled carbon nanotube networks under cyclic loading10.1007/s12274-011-0169-y ↗
11. Gupta · 2024
12. Temperature effect on the growth of carbon nanotubes using thermal chemical vapor deposition10.1016/S0009-2614(01)00680-7 ↗
13. Number of walls controlled synthesis of millimeter-long vertically aligned brushlike carbon nanotubes10.1021/jp0666986 ↗
14. Influence of ethylene and hydrogen flow rates on the wall number, crystallinity, and length of millimeter-long carbon nanotube array10.1021/jp802998v ↗
15. Tailoring the microstructure and mechanical properties of arrays of aligned multiwall carbon nanotubes by utilizing different hydrogen concentrations during synthesis10.1016/j.carbon.2011.04.066 ↗
16. Diverse 3D microarchitectures made by capillary forming of carbon nanotubes10.1002/adma.201001893 ↗
17. Fabrication and electrical integration of robust carbon nanotube micropillars by self-directed elastocapillary densification10.1088/0960-1317/21/4/045033 ↗
18. Geometry-induced mechanical properties of carbon nanotube foams10.1002/adem.201300524 ↗
19. Dynamic behavior of vertically aligned carbon nanotube foams with patterned microstructure10.1002/adem.201400571 ↗
20. Superior mechanical properties by exploiting size-effects and multiscale interactions in hierarchically architected foams10.1016/j.eml.2022.101899 ↗
21. Disrupting density-dependent property scaling in hierarchically architected foams10.1021/acsnano.3c01223 ↗
22. Moment invariants for multi-component shapes with applications to leaf classification10.1016/j.compag.2017.08.029 ↗
23. Disconnectedness: a new moment invariant for multi-component shapes10.1016/j.patcog.2018.01.010 ↗
24. Single-and multi-wall carbon nanotubes produced using the floating catalyst method: Synthesis, purification and hydrogen up-take10.1016/j.carbon.2006.03.009 ↗
25. Superior thermal transport properties of vertically aligned carbon nanotubes tailored through mesoscale architectures10.1016/j.carbon.2023.118526 ↗
26. Visual pattern recognition by moment invariants10.1109/TIT.1962.1057692 ↗
27. Artificial intelligence and machine learning in design of mechanical materials10.1039/D0MH01451F ↗
28. Thermodynamics-based artificial neural networks for constitutive modeling10.1016/j.jmps.2020.104277 ↗
29. Accelerated design and characterization of non-uniform cellular materials via a machine-learning based framework10.1038/s41524-020-0309-6 ↗
30. A machine learning approach to fracture mechanics problems10.1016/j.actamat.2020.03.016 ↗
31. Structure-mechanical stability relations of metal-organic frameworks via machine learning10.1016/j.matt.2019.03.002 ↗
32. Prediction and optimization of mechanical properties of composites using convolutional neural networks10.1016/j.compstruct.2019.111264 ↗
33. Prediction of composite microstructure stress-strain curves using convolutional neural networks10.1016/j.matdes.2020.108509 ↗
34. State-of-the-art in artificial neural network applications: A survey10.1016/j.heliyon.2018.e00938 ↗
35. Multilayer perceptron (MLP)
36. The Levenberg-Marquardt algorithm: implementation and theory
37. Bayesian regularization of neural networks
38. Predictive abilities of Bayesian regularization and Levenberg–Marquardt algorithms in artificial neural networks: a comparative empirical study on social data
39. Design of neural network with Levenberg-Marquardt and Bayesian regularization backpropagation for solving pantograph delay differential equations10.1109/ACCESS.2020.3011820 ↗
40. Comparative analysis of Bayesian regularization and Levenberg-Marquardt training algorithm for localization in wireless sensor network
41. Understanding deep learning (still) requires rethinking generalization10.1145/3446776 ↗
42. Mitigating overfitting in deep learning: Insights from Bayesian regularization
43. Application of artificial neural networks for predicting the bearing capacity of shallow foundations on rock masses10.1007/s00603-021-02549-1 ↗
44. Deep learning method for predicting the mechanical properties of aluminum alloys with small data sets
45. A generalized machine learning framework to estimate fatigue life across materials with minimal data10.1016/j.matdes.2024.113355 ↗
46. Data splitting for artificial neural networks using SOM-based stratified sampling10.1016/j.neunet.2009.11.009 ↗
47. Sensitivity analysis of the artificial neural networks in a system for durability prediction of forging tools to forgings made of C45 steel10.1007/s00170-020-05641-y ↗
48. Biomedical applications of soft robotics10.1038/s41578-018-0022-y ↗