Geometrical shape learning as basis for compact microstructure representations and microstructure-properties linkages
Process-structure-properties linkages play a major role in materials and process engineering. Nowadays, such linkages are often established on the basis of experimental data and simulation data using machine learning approaches. For this purpose, typically, state-of-the-art feature extraction method...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | European Journal of Materials |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/26889277.2025.2454654 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832557393352851456 |
|---|---|
| author | Rodrigo Iza Teran Daniela Steffes-lai Lukas Morand |
| author_facet | Rodrigo Iza Teran Daniela Steffes-lai Lukas Morand |
| author_sort | Rodrigo Iza Teran |
| collection | DOAJ |
| description | Process-structure-properties linkages play a major role in materials and process engineering. Nowadays, such linkages are often established on the basis of experimental data and simulation data using machine learning approaches. For this purpose, typically, state-of-the-art feature extraction methods, such as principal component analysis, are used in combination with regression models, such as neural networks. For complex spatially-resolved microstructure representations convolutional neural networks are often used, which are, however, very data-intensive and not explainable. In this work, we present a novel approach based on geometrical shape features that allows for compact microstructure representations, even with a small amount of data, and is, furthermore, explainable. In addition, the presented approach maps the identified features to a latent feature space that is not dependent on the data. |
| format | Article |
| id | doaj-art-a1e642e056f74e7b901f6b742bfcc749 |
| institution | Kabale University |
| issn | 2688-9277 |
| language | English |
| publishDate | 2025-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | European Journal of Materials |
| spelling | doaj-art-a1e642e056f74e7b901f6b742bfcc7492025-02-03T05:12:09ZengTaylor & Francis GroupEuropean Journal of Materials2688-92772025-12-015110.1080/26889277.2025.2454654Geometrical shape learning as basis for compact microstructure representations and microstructure-properties linkagesRodrigo Iza Teran0Daniela Steffes-lai1Lukas Morand2Fraunhofer Institute for Algorithms and Scientific Computing SCAI, Sankt Augustin, GermanyFraunhofer Institute for Algorithms and Scientific Computing SCAI, Sankt Augustin, GermanyFraunhofer Institute for Mechanics of Materials IWM, Freiburg, GermanyProcess-structure-properties linkages play a major role in materials and process engineering. Nowadays, such linkages are often established on the basis of experimental data and simulation data using machine learning approaches. For this purpose, typically, state-of-the-art feature extraction methods, such as principal component analysis, are used in combination with regression models, such as neural networks. For complex spatially-resolved microstructure representations convolutional neural networks are often used, which are, however, very data-intensive and not explainable. In this work, we present a novel approach based on geometrical shape features that allows for compact microstructure representations, even with a small amount of data, and is, furthermore, explainable. In addition, the presented approach maps the identified features to a latent feature space that is not dependent on the data.https://www.tandfonline.com/doi/10.1080/26889277.2025.2454654Machine learningmicrostructure-properties linkagemicrostructure representationdimension reduction |
| spellingShingle | Rodrigo Iza Teran Daniela Steffes-lai Lukas Morand Geometrical shape learning as basis for compact microstructure representations and microstructure-properties linkages European Journal of Materials Machine learning microstructure-properties linkage microstructure representation dimension reduction |
| title | Geometrical shape learning as basis for compact microstructure representations and microstructure-properties linkages |
| title_full | Geometrical shape learning as basis for compact microstructure representations and microstructure-properties linkages |
| title_fullStr | Geometrical shape learning as basis for compact microstructure representations and microstructure-properties linkages |
| title_full_unstemmed | Geometrical shape learning as basis for compact microstructure representations and microstructure-properties linkages |
| title_short | Geometrical shape learning as basis for compact microstructure representations and microstructure-properties linkages |
| title_sort | geometrical shape learning as basis for compact microstructure representations and microstructure properties linkages |
| topic | Machine learning microstructure-properties linkage microstructure representation dimension reduction |
| url | https://www.tandfonline.com/doi/10.1080/26889277.2025.2454654 |
| work_keys_str_mv | AT rodrigoizateran geometricalshapelearningasbasisforcompactmicrostructurerepresentationsandmicrostructurepropertieslinkages AT danielasteffeslai geometricalshapelearningasbasisforcompactmicrostructurerepresentationsandmicrostructurepropertieslinkages AT lukasmorand geometricalshapelearningasbasisforcompactmicrostructurerepresentationsandmicrostructurepropertieslinkages |