Microstructure Estimation by Combining Deep Learning and Phase Transformation Model

被引:0
作者
Noguchi, Satoshi [1 ]
Aihara, Syuji [2 ]
Inoue, Junya [3 ,4 ]
机构
[1] Japan Agcy Marine Earth Sci & Technol, Res Inst Value Added Informat Generat, Yokosuka, Kanagawa, Japan
[2] Univ Tokyo, Tokyo, Japan
[3] Univ Tokyo, Inst Ind Sci, Tokyo, Japan
[4] Univ Tokyo, Dept Mat Engn, Tokyo, Japan
来源
TETSU TO HAGANE-JOURNAL OF THE IRON AND STEEL INSTITUTE OF JAPAN | 2023年 / 109卷 / 11期
关键词
microstructure estimation; deep learning; vector quantized variational autoencoder; pixel convolutional neural network; STRUCTURE-PROPERTY LINKAGES; HIGH-CONTRAST COMPOSITES; FIELD MODEL; RECONSTRUCTION; SOLIDIFICATION; PREDICTION;
D O I
10.2355/tetsutohagane.TETSU-2023-045
中图分类号
TF [冶金工业];
学科分类号
0806 ;
摘要
In material design, the establishment of process structure property relationship is crucial for analyzing and controlling material microstructures. For the establishment of process structure property relationship, a central problem is the analysis, characterization, and control of microstructures, since microstructures are highly sensitive to material processing and critically affect material's properties. Therefore, accurately estimating the morphology of material microstructures plays a significant role in understanding the process structure property relationship. In this paper, we propose a deep -learning framework for estimating material microstructures under specific process conditions. The framework utilizes two deep learning networks: Vector Quantized Variational Autoencoder (VQVAE) and Pixel Convolutional Neural Network (PixeICNN). The framework can predict material micrographs from the transformation behavior given by some physical model. In this sense, the framework is consistent with the physical knowledge accumulated in the field of material science. Importantly our study demonstrates qualitative and quantitative evidences that incorporating physical models enhances the accuracy of microstructure estimation by deep learning models. These results highlight the importance of appropriately integrating field -specific knowledge when applying data-driven frameworks to materials design. Consequently, our results provide a foundation for integrating data-driven methods with the accumulated knowledge in the field. This integration holds great potential for advancing material design through deep learning.
引用
收藏
页码:898 / 914
页数:17
相关论文
共 50 条
[21]   Combining a deep learning model with multivariate empirical mode decomposition for hourly global horizontal irradiance forecasting [J].
Gupta, Priya ;
Singh, Rhythm .
RENEWABLE ENERGY, 2023, 206 (908-927) :908-927
[22]   Combining CERES-Wheat model, Sentinel-2 data, and deep learning method for winter wheat yield estimation [J].
Xie, Yi .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 2022, 43 (02) :630-648
[23]   Estimation of missing values in heterogeneous traffic data: Application of multimodal deep learning model [J].
Li, Linchao ;
Du, Bowen ;
Wang, Yonggang ;
Qin, Lingqiao ;
Tan, Huachun .
KNOWLEDGE-BASED SYSTEMS, 2020, 194
[24]   Hybrid deep learning model for concrete incorporating microencapsulated phase change materials [J].
Tanyildizi, Harun ;
Marani, Afshin ;
Turk, Kazim ;
Nehdi, Moncef L. .
CONSTRUCTION AND BUILDING MATERIALS, 2022, 319
[25]   Microstructure Recognition by Deep Learning [J].
Adachi, Yoshitaka ;
Taguchi, Motoki ;
Hirokawa, Shogo .
TETSU TO HAGANE-JOURNAL OF THE IRON AND STEEL INSTITUTE OF JAPAN, 2016, 102 (12) :722-729
[26]   DGImNet: A deep learning model for photovoltaic soiling loss estimation [J].
Fang, Mingyu ;
Qian, Weixing ;
Qian, Tao ;
Bao, Qiwei ;
Zhang, Haocheng ;
Qiu, Xiao .
APPLIED ENERGY, 2024, 376
[27]   ModelRevelator: Fast phylogenetic model estimation via deep learning [J].
Burgstaller-Muehlbacher, Sebastian ;
Crotty, Stephen M. ;
Schmidt, Heiko A. ;
Reden, Franziska ;
Drucks, Tamara ;
von Haeseler, Arndt .
MOLECULAR PHYLOGENETICS AND EVOLUTION, 2023, 188
[28]   An Integrated Model Combining Machine Learning and Deep Learning Algorithms for Classification of Rupture Status of IAs [J].
Chen, Rong ;
Mo, Xiao ;
Chen, Zhenpeng ;
Feng, Pujie ;
Li, Haiyun .
FRONTIERS IN NEUROLOGY, 2022, 13
[29]   Research on Vehicle Distance Estimation Model based on Deep Learning [J].
Ding, Cheng ;
Bao, Jianmin ;
Mi, Guanyu ;
Kuai, Xiao ;
Kang, YiNing .
2022 41ST CHINESE CONTROL CONFERENCE (CCC), 2022, :7169-7173
[30]   A microstructure model for recrystallisation and phase transformation during the dual-phase steel annealing cycle [J].
Bos, C. ;
Mecozzi, M. G. ;
Sietsma, J. .
COMPUTATIONAL MATERIALS SCIENCE, 2010, 48 (03) :692-699