Harnessing structural stochasticity in the computational discovery and design of microstructures
This paper presents a deep generative model-based design methodology for tailoring the structural stochasticity of microstructures. Although numerous methods have been established for designing deterministic (periodic) or stochastic microstructures, a systematic design approach that allows the unifi...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2022-11-01
|
Series: | Materials & Design |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S0264127522008450 |
_version_ | 1811249607640875008 |
---|---|
author | Leidong Xu Nathaniel Hoffman Zihan Wang Hongyi Xu |
author_facet | Leidong Xu Nathaniel Hoffman Zihan Wang Hongyi Xu |
author_sort | Leidong Xu |
collection | DOAJ |
description | This paper presents a deep generative model-based design methodology for tailoring the structural stochasticity of microstructures. Although numerous methods have been established for designing deterministic (periodic) or stochastic microstructures, a systematic design approach that allows the unified treatment of both deterministic and stochastic microstructure design domains has yet to be created. The proposed methodology resolves this issue by learning a unified feature space that embodies diverse structural patterns with continuously varying stochasticity levels. A highly diverse microstructure database is established to incorporate various types of deterministic and stochastic microstructure patterns. A property-aware deep generative model is proposed to learn a unified feature space of the structural characteristics, as well as the relationship between structure features and properties of interest. Autoencoder (AE), Variational Autoencoder (VAE), and Adversarial Autoencoder (AAE) are compared to understand their relative merits in the property-aware learning of the unified feature space. Microstructural designs with tailorable stochasticity and properties are obtained by searching the unified feature space. Multiple design cases are presented to demonstrate the capability of designing microstructures for structural stochasticity and properties. Furthermore, the proposed method is employed to create stochastically graded structures, which manipulate the mechanical behaviors by varying the local stochasticity of the structure. |
first_indexed | 2024-04-12T15:49:33Z |
format | Article |
id | doaj.art-48201372304e45f39346a3d9567abe0a |
institution | Directory Open Access Journal |
issn | 0264-1275 |
language | English |
last_indexed | 2024-04-12T15:49:33Z |
publishDate | 2022-11-01 |
publisher | Elsevier |
record_format | Article |
series | Materials & Design |
spelling | doaj.art-48201372304e45f39346a3d9567abe0a2022-12-22T03:26:33ZengElsevierMaterials & Design0264-12752022-11-01223111223Harnessing structural stochasticity in the computational discovery and design of microstructuresLeidong Xu0Nathaniel Hoffman1Zihan Wang2Hongyi Xu3Department of Mechanical Engineering, University of Connecticut, 191 Auditorium Road, Storrs, CT 06269, United StatesDepartment of Mechanical Engineering, University of Connecticut, 191 Auditorium Road, Storrs, CT 06269, United StatesDepartment of Mechanical Engineering, University of Connecticut, 191 Auditorium Road, Storrs, CT 06269, United StatesCorresponding author.; Department of Mechanical Engineering, University of Connecticut, 191 Auditorium Road, Storrs, CT 06269, United StatesThis paper presents a deep generative model-based design methodology for tailoring the structural stochasticity of microstructures. Although numerous methods have been established for designing deterministic (periodic) or stochastic microstructures, a systematic design approach that allows the unified treatment of both deterministic and stochastic microstructure design domains has yet to be created. The proposed methodology resolves this issue by learning a unified feature space that embodies diverse structural patterns with continuously varying stochasticity levels. A highly diverse microstructure database is established to incorporate various types of deterministic and stochastic microstructure patterns. A property-aware deep generative model is proposed to learn a unified feature space of the structural characteristics, as well as the relationship between structure features and properties of interest. Autoencoder (AE), Variational Autoencoder (VAE), and Adversarial Autoencoder (AAE) are compared to understand their relative merits in the property-aware learning of the unified feature space. Microstructural designs with tailorable stochasticity and properties are obtained by searching the unified feature space. Multiple design cases are presented to demonstrate the capability of designing microstructures for structural stochasticity and properties. Furthermore, the proposed method is employed to create stochastically graded structures, which manipulate the mechanical behaviors by varying the local stochasticity of the structure.http://www.sciencedirect.com/science/article/pii/S0264127522008450Microstructure designDeep generative modelStochasticityDesign optimizationStochastically graded structures |
spellingShingle | Leidong Xu Nathaniel Hoffman Zihan Wang Hongyi Xu Harnessing structural stochasticity in the computational discovery and design of microstructures Materials & Design Microstructure design Deep generative model Stochasticity Design optimization Stochastically graded structures |
title | Harnessing structural stochasticity in the computational discovery and design of microstructures |
title_full | Harnessing structural stochasticity in the computational discovery and design of microstructures |
title_fullStr | Harnessing structural stochasticity in the computational discovery and design of microstructures |
title_full_unstemmed | Harnessing structural stochasticity in the computational discovery and design of microstructures |
title_short | Harnessing structural stochasticity in the computational discovery and design of microstructures |
title_sort | harnessing structural stochasticity in the computational discovery and design of microstructures |
topic | Microstructure design Deep generative model Stochasticity Design optimization Stochastically graded structures |
url | http://www.sciencedirect.com/science/article/pii/S0264127522008450 |
work_keys_str_mv | AT leidongxu harnessingstructuralstochasticityinthecomputationaldiscoveryanddesignofmicrostructures AT nathanielhoffman harnessingstructuralstochasticityinthecomputationaldiscoveryanddesignofmicrostructures AT zihanwang harnessingstructuralstochasticityinthecomputationaldiscoveryanddesignofmicrostructures AT hongyixu harnessingstructuralstochasticityinthecomputationaldiscoveryanddesignofmicrostructures |