2025-08-11 デューク大学（Duke）
The cubes’ internal structures are based on different porous materials, including bone and wood.
＜関連情報＞
- https://pratt.duke.edu/news/predictions-under-pressure-using-ai-to-study-porous-materials/
- https://www.nature.com/articles/s44172-025-00410-9
- https://pubs.acs.org/doi/10.1021/acsomega.5c00641
- https://arxiv.org/abs/2501.10481
形態情報に基づく神経ネットワークを用いて、弾塑性多孔質媒体の圧縮応力-ひずみ挙動を予測する Predicting compressive stress-strain behavior of elasto-plastic porous media via morphology-informed neural networks
W. Lindqwister,J. Peloquin,L. E. Dalton,K. Gall & M. Veveakis
Communications Engineering Published:18 April 2025
DOI:https://doi.org/10.1038/s44172-025-00410-9
Abstract
Porous media, ranging from bones to concrete and from batteries to architected lattices, pose difficult challenges in fully harnessing for engineering applications due to their complex and variable structures. Accurate and rapid assessment of their mechanical behavior is both challenging and essential, and traditional methods such as destructive testing and finite element analysis can be costly, computationally demanding, and time consuming. Machine learning (ML) offers a promising alternative for predicting mechanical behavior by leveraging data-driven correlations. However, with such structural complexity and diverse morphology among porous media, the question becomes how to effectively characterize these materials to provide robust feature spaces for ML that are descriptive, succinct, and easily interpreted. Here, we developed an automated methodology to determine porous material strength. This method uses scalar morphological descriptors, known as Minkowski functionals, to describe the porous space. From there, we conduct uniaxial compression experiments for generating material stress-strain curves, and then train an ML model to predict the curves using said morphological descriptors. This framework seeks to expedite the analysis and prediction of stress-strain behavior in porous materials and lay the groundwork for future models that can predict mechanical behaviors beyond compression.
多孔質媒体における非混合反応界面のための化学的均一化 Chemical Homogenization for Nonmixing Reactive Interfaces in Porous Media
Winston Lindqwister,Manolis Veveakis,and Martin Lesueur
ACS Omega Published: May 21, 2025
DOI:https://doi.org/10.1021/acsomega.5c00641
Abstract
Through rocks and concrete, batteries, and bone, porous media represent a wide class of materials whose chemical makeup and reactivity directly impact their behavior at multiple scales. While various theoretical and computational models have been implemented to capture the chemical behavior of these systems, none have investigated how the very geometry of porous media, the structures that make these materials porous and define the interfaces between solids and fluids, affects these behaviors. Through this work, we explored Minkowski functionals–geometric morphometers that describe the spatial and topological features of a convex space–to investigate how microstructural morphology affects systemic chemical performance. Using a novel asynchronous cellular automaton known as a surface chemical reaction network (CRN) to model chemical behavior, linkages were found between Minkowski functionals and equilibrium constant, as well as properties related to the dynamics of the microstructure’s reaction quotient. These quantities, in turn, give insight into how morphology affects bulk porous media properties, such as Gibbs’ free energy.
潜在的硬化学習（LLH）：材料逆問題におけるドメイン知識を活用した深層学習の強化 Learning Latent Hardening (LLH): Enhancing Deep Learning with Domain Knowledge for Material Inverse Problems
Qinyi Tian, Winston Lindqwister, Manolis Veveakis, Laura E. Dalton
arXiv last revised 9 Apr 2025 (this version, v3)
DOI:https://doi.org/10.48550/arXiv.2501.10481
Abstract
Advancements in deep learning and machine learning have improved the ability to model complex, nonlinear relationships, such as those encountered in complex material inverse problems. However, the effectiveness of these methods often depends on large datasets, which are not always available. In this study, the incorporation of domain-specific knowledge of the mechanical behavior of material microstructures is investigated to evaluate the impact on the predictive performance of the models in data-scarce scenarios. To overcome data limitations, a two-step framework, Learning Latent Hardening (LLH), is proposed. In the first step of LLH, a Deep Neural Network is employed to reconstruct full stress-strain curves from randomly selected portions of the stress-strain curves to capture the latent mechanical response of a material based on key microstructural features. In the second step of LLH, the results of the reconstructed stress-strain curves are leveraged to predict key microstructural features of porous materials. The performance of six deep learning and/or machine learning models trained with and without domain knowledge are compared: Convolutional Neural Networks, Deep Neural Networks, Extreme Gradient Boosting, K-Nearest Neighbors, Long Short-Term Memory, and Random Forest. The results from the models with domain-specific information consistently achieved higher R 2 values compared to models without prior knowledge. Models without domain knowledge missed critical patterns linking stress-strain behavior to microstructural changes, whereas domain-informed models better identified essential stress-strain features predictive of microstructure. These findings highlight the importance of integrating domain-specific knowledge with deep learning to achieve accurate outcomes in materials science.
