Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder
Data-driven approaches enable a deep understanding of microstructure and mechanical properties of materials and greatly promote one's capability in designing new advanced materials. Deep learning-based image processing outperforms conventional image processing techniques with unsupervised learn...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-04-01
|
| Series: | Materials & Design |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0264127521000976 |
| _version_ | 1818675943765442560 |
|---|---|
| author | Yongju Kim Hyung Keun Park Jaimyun Jung Peyman Asghari-Rad Seungchul Lee Jin You Kim Hwan Gyo Jung Hyoung Seop Kim |
| author_facet | Yongju Kim Hyung Keun Park Jaimyun Jung Peyman Asghari-Rad Seungchul Lee Jin You Kim Hwan Gyo Jung Hyoung Seop Kim |
| author_sort | Yongju Kim |
| collection | DOAJ |
| description | Data-driven approaches enable a deep understanding of microstructure and mechanical properties of materials and greatly promote one's capability in designing new advanced materials. Deep learning-based image processing outperforms conventional image processing techniques with unsupervised learning. This study employs a variational autoencoder (VAE) to generate a continuous microstructure space based on synthetic microstructural images. The structure-property relationships are explored using a computational approach with microstructure quantification, dimensionality reduction, and finite element method (FEM) simulations. The FEM of representative volume element (RVE) with a microstructure-based constitutive model model is proposed for predicting the overall stress-strain behavior of the investigated dual-phase steels. Then, Gaussian process regression (GPR) is used to make connections between the latent space point and the ferrite grain size as inputs and mechanical properties as outputs. The GPR with VAE successfully predicts the newly generated microstructures with target mechanical properties with high accuracy. This work demonstrates that a variety of microstructures can be candidates for designing the optimal material with target properties in a continuous manner. |
| first_indexed | 2024-12-17T08:35:37Z |
| format | Article |
| id | doaj.art-fd9eb26e4d2c45f28156968b2b8a8756 |
| institution | Directory Open Access Journal |
| issn | 0264-1275 |
| language | English |
| last_indexed | 2024-12-17T08:35:37Z |
| publishDate | 2021-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Materials & Design |
| spelling | doaj.art-fd9eb26e4d2c45f28156968b2b8a87562022-12-21T21:56:29ZengElsevierMaterials & Design0264-12752021-04-01202109544Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoderYongju Kim0Hyung Keun Park1Jaimyun Jung2Peyman Asghari-Rad3Seungchul Lee4Jin You Kim5Hwan Gyo Jung6Hyoung Seop Kim7Department of Materials Science and Engineering, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of KoreaDepartment of Materials Science and Engineering, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of KoreaDepartment of Virtual Materials Processing, Korea Institute of Materials Science, Changwon 51508, Republic of KoreaDepartment of Materials Science and Engineering, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea; Center for High Entropy Alloys, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of KoreaDepartment of Mechanical Engineering, Pohang University of Science and Technology (POSTECH), Pohang, 37673, Republic of KoreaTechnical Research Lab. Energy and Shipbuilding Steel Research Group, POSCO, Pohang 790-785, Republic of KoreaTechnical Research Lab. Energy and Shipbuilding Steel Research Group, POSCO, Pohang 790-785, Republic of KoreaDepartment of Materials Science and Engineering, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea; Center for High Entropy Alloys, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea; Graduate Institute of Ferrous Technology Pohang University of Science and Technology (POSTECH), Pohang, 37673, Republic of Korea; Corresponding author at: Graduate Institute of Ferrous Technology Pohang University of Science and Technology (POSTECH), Pohang, 37673, Republic of Korea.Data-driven approaches enable a deep understanding of microstructure and mechanical properties of materials and greatly promote one's capability in designing new advanced materials. Deep learning-based image processing outperforms conventional image processing techniques with unsupervised learning. This study employs a variational autoencoder (VAE) to generate a continuous microstructure space based on synthetic microstructural images. The structure-property relationships are explored using a computational approach with microstructure quantification, dimensionality reduction, and finite element method (FEM) simulations. The FEM of representative volume element (RVE) with a microstructure-based constitutive model model is proposed for predicting the overall stress-strain behavior of the investigated dual-phase steels. Then, Gaussian process regression (GPR) is used to make connections between the latent space point and the ferrite grain size as inputs and mechanical properties as outputs. The GPR with VAE successfully predicts the newly generated microstructures with target mechanical properties with high accuracy. This work demonstrates that a variety of microstructures can be candidates for designing the optimal material with target properties in a continuous manner.http://www.sciencedirect.com/science/article/pii/S0264127521000976Microstructure-based modelingDeep learningVariational autoencoderGaussian process regressionDual-phase steel |
| spellingShingle | Yongju Kim Hyung Keun Park Jaimyun Jung Peyman Asghari-Rad Seungchul Lee Jin You Kim Hwan Gyo Jung Hyoung Seop Kim Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder Materials & Design Microstructure-based modeling Deep learning Variational autoencoder Gaussian process regression Dual-phase steel |
| title | Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder |
| title_full | Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder |
| title_fullStr | Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder |
| title_full_unstemmed | Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder |
| title_short | Exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder |
| title_sort | exploration of optimal microstructure and mechanical properties in continuous microstructure space using a variational autoencoder |
| topic | Microstructure-based modeling Deep learning Variational autoencoder Gaussian process regression Dual-phase steel |
| url | http://www.sciencedirect.com/science/article/pii/S0264127521000976 |
| work_keys_str_mv | AT yongjukim explorationofoptimalmicrostructureandmechanicalpropertiesincontinuousmicrostructurespaceusingavariationalautoencoder AT hyungkeunpark explorationofoptimalmicrostructureandmechanicalpropertiesincontinuousmicrostructurespaceusingavariationalautoencoder AT jaimyunjung explorationofoptimalmicrostructureandmechanicalpropertiesincontinuousmicrostructurespaceusingavariationalautoencoder AT peymanasgharirad explorationofoptimalmicrostructureandmechanicalpropertiesincontinuousmicrostructurespaceusingavariationalautoencoder AT seungchullee explorationofoptimalmicrostructureandmechanicalpropertiesincontinuousmicrostructurespaceusingavariationalautoencoder AT jinyoukim explorationofoptimalmicrostructureandmechanicalpropertiesincontinuousmicrostructurespaceusingavariationalautoencoder AT hwangyojung explorationofoptimalmicrostructureandmechanicalpropertiesincontinuousmicrostructurespaceusingavariationalautoencoder AT hyoungseopkim explorationofoptimalmicrostructureandmechanicalpropertiesincontinuousmicrostructurespaceusingavariationalautoencoder |