Cargando…

Martensite Start Temperature Prediction through a Deep Learning Strategy Using Both Microstructure Images and Composition Data

In recent decades, various previous research has established empirical formulae or thermodynamic models for martensite start temperature (Ms) prediction. However, most of this research has mainly considered the effect of composition and ignored complex microstructural factors, such as morphology, th...

Descripción completa

Detalles Bibliográficos
Autores principales: Yang, Zenan, Li, Yong, Wei, Xiaolu, Wang, Xu, Wang, Chenchong
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9917892/
https://www.ncbi.nlm.nih.gov/pubmed/36769939
http://dx.doi.org/10.3390/ma16030932
_version_ 1784886477550780416
author Yang, Zenan
Li, Yong
Wei, Xiaolu
Wang, Xu
Wang, Chenchong
author_facet Yang, Zenan
Li, Yong
Wei, Xiaolu
Wang, Xu
Wang, Chenchong
author_sort Yang, Zenan
collection PubMed
description In recent decades, various previous research has established empirical formulae or thermodynamic models for martensite start temperature (Ms) prediction. However, most of this research has mainly considered the effect of composition and ignored complex microstructural factors, such as morphology, that significantly affect Ms. The main limitation is that most microstructures cannot be digitized into numerical data. In order to solve this problem, a convolutional neural network model that can use both composition information and microstructure images as input was established for Ms prediction in a medium-Mn steel system in this research. Firstly, the database was established through experimenting. Then, the model was built and trained with the database. Finally, the performance of the model was systematically evaluated based on comparison with other, traditional AI models. It was proven that the new model provided in this research is more rational and accurate because it considers both composition and microstructural factors. In addition, because of the use of microstructure images for data augmentation, the deep learning had a low risk of overfitting. When the deep-learning strategy is used to deal with data that contains both numerical and image data types, obtaining the value matrix that contains interaction information of both numerical and image data through data preprocessing is probably a better approach than direct linking of the numerical data vector to the fully connected layer.
format Online
Article
Text
id pubmed-9917892
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-99178922023-02-11 Martensite Start Temperature Prediction through a Deep Learning Strategy Using Both Microstructure Images and Composition Data Yang, Zenan Li, Yong Wei, Xiaolu Wang, Xu Wang, Chenchong Materials (Basel) Article In recent decades, various previous research has established empirical formulae or thermodynamic models for martensite start temperature (Ms) prediction. However, most of this research has mainly considered the effect of composition and ignored complex microstructural factors, such as morphology, that significantly affect Ms. The main limitation is that most microstructures cannot be digitized into numerical data. In order to solve this problem, a convolutional neural network model that can use both composition information and microstructure images as input was established for Ms prediction in a medium-Mn steel system in this research. Firstly, the database was established through experimenting. Then, the model was built and trained with the database. Finally, the performance of the model was systematically evaluated based on comparison with other, traditional AI models. It was proven that the new model provided in this research is more rational and accurate because it considers both composition and microstructural factors. In addition, because of the use of microstructure images for data augmentation, the deep learning had a low risk of overfitting. When the deep-learning strategy is used to deal with data that contains both numerical and image data types, obtaining the value matrix that contains interaction information of both numerical and image data through data preprocessing is probably a better approach than direct linking of the numerical data vector to the fully connected layer. MDPI 2023-01-18 /pmc/articles/PMC9917892/ /pubmed/36769939 http://dx.doi.org/10.3390/ma16030932 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Yang, Zenan
Li, Yong
Wei, Xiaolu
Wang, Xu
Wang, Chenchong
Martensite Start Temperature Prediction through a Deep Learning Strategy Using Both Microstructure Images and Composition Data
title Martensite Start Temperature Prediction through a Deep Learning Strategy Using Both Microstructure Images and Composition Data
title_full Martensite Start Temperature Prediction through a Deep Learning Strategy Using Both Microstructure Images and Composition Data
title_fullStr Martensite Start Temperature Prediction through a Deep Learning Strategy Using Both Microstructure Images and Composition Data
title_full_unstemmed Martensite Start Temperature Prediction through a Deep Learning Strategy Using Both Microstructure Images and Composition Data
title_short Martensite Start Temperature Prediction through a Deep Learning Strategy Using Both Microstructure Images and Composition Data
title_sort martensite start temperature prediction through a deep learning strategy using both microstructure images and composition data
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9917892/
https://www.ncbi.nlm.nih.gov/pubmed/36769939
http://dx.doi.org/10.3390/ma16030932
work_keys_str_mv AT yangzenan martensitestarttemperaturepredictionthroughadeeplearningstrategyusingbothmicrostructureimagesandcompositiondata
AT liyong martensitestarttemperaturepredictionthroughadeeplearningstrategyusingbothmicrostructureimagesandcompositiondata
AT weixiaolu martensitestarttemperaturepredictionthroughadeeplearningstrategyusingbothmicrostructureimagesandcompositiondata
AT wangxu martensitestarttemperaturepredictionthroughadeeplearningstrategyusingbothmicrostructureimagesandcompositiondata
AT wangchenchong martensitestarttemperaturepredictionthroughadeeplearningstrategyusingbothmicrostructureimagesandcompositiondata