Fast and accurate prediction of temperature evolutions in additive manufacturing process using deep learning

被引:0
|
作者
Thinh Quy Duc Pham
Truong Vinh Hoang
Xuan Van Tran
Quoc Tuan Pham
Seifallah Fetni
Laurent Duchêne
Hoang Son Tran
Anne-Marie Habraken
机构
[1] Thu Dau Mot University,Institute of Strategy Development
[2] RWTH-Aachen University,Chair of Mathematics for Uncertainty Quantification
[3] Ton Duc Thang University,Division of Computational Mathematics and Engineering, Institute for Computational Science
[4] Ton Duc Thang University,Faculty of Civil Engineering
[5] University of Liège,undefined
[6] Fonds de la Recherche Scientifique de Belgique (F.R.S-FNRS),undefined
来源
Journal of Intelligent Manufacturing | 2023年 / 34卷
关键词
Deep learning; Directed energy deposition; Temperature evolutions; Sensitivity analysis; SHAP method;
D O I
暂无
中图分类号
学科分类号
摘要
Typical computer-based parameter optimization and uncertainty quantification of the additive manufacturing process usually requires significant computational cost for performing high-fidelity heat transfer finite element (FE) models with different process settings. This work develops a simple surrogate model using a feedforward neural network (FFNN) for a fast and accurate prediction of the temperature evolutions and the melting pool sizes in a metal bulk sample (3D horizontal layers) manufactured by the DED process. Our surrogate model is trained using high-fidelity data obtained from the FE model, which was validated by experiments. The temperature evolutions and the melting pool sizes predicted by the FFNN model exhibit accuracy of 99%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$99\%$$\end{document} and 98%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$98\%$$\end{document}, respectively, compared with the FE model for unseen process settings in the studied range. Moreover, to evaluate the importance of the input features and explain the achieved accuracy of the FFNN model, a sensitivity analysis (SA) is carried out using the SHapley Additive exPlanation (SHAP) method. The SA shows that the most critical enriched features impacting the predictive capability of the FFNN model are the vertical distance from the laser head position to the material point and the laser head position.
引用
收藏
页码:1701 / 1719
页数:18
相关论文
共 50 条
  • [21] Surface Roughness Prediction in Additive Manufacturing Using Machine Learning
    Wu, Dazhong
    Wei, Yupeng
    Terpenny, Janis
    PROCEEDINGS OF THE ASME 13TH INTERNATIONAL MANUFACTURING SCIENCE AND ENGINEERING CONFERENCE, 2018, VOL 3, 2018,
  • [22] Fast and Accurate MEG Source Localization using Deep Learning
    Wang, Hanchen
    Feng, Shihang
    Zhang, Qian
    Kim, Young Jin
    Savukov, Igor
    Yang, Lan
    Lin, Youzuo
    MEDICAL IMAGING 2024: PHYSICS OF MEDICAL IMAGING, PT 1, 2024, 12925
  • [23] Fast and accurate view classification of echocardiograms using deep learning
    Ali Madani
    Ramy Arnaout
    Mohammad Mofrad
    Rima Arnaout
    npj Digital Medicine, 1
  • [24] Fast and accurate view classification of echocardiograms using deep learning
    Madani, Ali
    Arnaout, Ramy
    Mofrad, Mohammad
    Arnaout, Rima
    NPJ DIGITAL MEDICINE, 2018, 1
  • [25] A hybrid deep learning model of process-build interactions in additive manufacturing
    Yazdi, Reza Mojahed
    Imani, Farhad
    Yang, Hui
    JOURNAL OF MANUFACTURING SYSTEMS, 2020, 57 : 460 - 468
  • [26] Fast and accurate prediction of failure pressure of oil and gas defective pipelines using the deep learning model
    Su, Yue
    Li, Jingfa
    Yu, Bo
    Zhao, Yanlin
    Yao, Jun
    RELIABILITY ENGINEERING & SYSTEM SAFETY, 2021, 216
  • [27] Object and defect detection in additive manufacturing using deep learning algorithms
    da Silva, Lucas Macedo
    Alcala, Symone G. S.
    Barbosa, Talles Marcelo G. de A.
    Araujo, Rui
    PRODUCTION ENGINEERING-RESEARCH AND DEVELOPMENT, 2024, 18 (06): : 889 - 902
  • [28] Highly accurate prediction of specific activity using deep learning
    Sheinfeld, Mati
    Levinson, Samuel
    Orion, Itzhak
    APPLIED RADIATION AND ISOTOPES, 2017, 130 : 115 - 120
  • [29] Accurate prediction of chromatin conformation status using deep learning
    Uryu, Hidetaka
    Hata, Kenichiro
    CANCER SCIENCE, 2018, 109 : 845 - 845
  • [30] Zoning additive manufacturing process histories using unsupervised machine learning
    Donegan, Sean P.
    Schwalbach, Edwin J.
    Groeber, Michael A.
    MATERIALS CHARACTERIZATION, 2020, 161 (161)