A Deep-Learning-Based Surrogate Model for Thermal Signature Prediction in Laser Metal Deposition

被引:28
作者
Guo, Shenghan [1 ]
Guo, Weihong [2 ]
Bian, Linkan [3 ]
Guo, Y. B. [4 ,5 ]
机构
[1] Arizona State Univ, Sch Mfg Syst & Networks, Mesa, AZ 85212 USA
[2] State Univ New Jersey, Dept Ind & Syst Engn, Rutgers, Piscataway, NJ 08854 USA
[3] Mississippi State Univ, Dept Ind & Syst Engn, Starkville, MS 39762 USA
[4] State Univ New Jersey, Dept Mech & Aerosp Engn, Rutgers, Piscataway, NJ 08854 USA
[5] State Univ New Jersey, New Jersey Adv Engn Inst, Rutgers, Piscataway, NJ 08854 USA
关键词
Transient analysis; Predictive models; Metals; Computational modeling; Data models; Numerical models; Training; Laser metal deposition; thermal images; surrogate model; generative adversarial net; physics-guided image generation; SIMULATION; FIELD;
D O I
10.1109/TASE.2022.3158204
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Laser metal deposition (LMD) is an additive manufacturing method for metal parts by using focused thermal energy to fuse materials as they are deposited. During LMD, transient thermal signatures such as the in-situ thermal images of melt pool, contain rich information about process performance. Early prediction of such transient thermal signatures provides opportunities for process monitoring and defect prevention. While physics-based models of LMD have been conventionally used for thermal signature prediction, they have limitations and are computationally expensive for real-time prediction. A scalable, efficient data-science-based model is therefore needed. This paper develops a deep-learning-based surrogate model, called LMD-cGAN, to predict and emulate the transient thermal signatures in LMD. The model generates images for the thermal dynamics of melt pool conditionally on the deposition layer. It enables early prediction of future-layer thermal signatures for an in-process part based on its early-layer thermal signatures. To respect the physics in LMD, a physics-guided image selection (PGIS) mechanism is integrated with LMD-cGAN to calibrate the predictions against physical benchmarks of transient melt pool for the process. The effectiveness and efficiency of the proposed method are demonstrated in a case study on the LMD of Ti-4Al-6V thin-walled structures.
引用
收藏
页码:482 / 494
页数:13
相关论文
共 52 条
[41]  
Tilton M., 2018, Orthopedic biomaterials: progress in biology, manufacturing, and industry perspectives, P21, DOI 10.1007/978-3-319-89542-02
[42]   High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs [J].
Wang, Ting-Chun ;
Liu, Ming-Yu ;
Zhu, Jun-Yan ;
Tao, Andrew ;
Kautz, Jan ;
Catanzaro, Bryan .
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :8798-8807
[43]   Uncertainty Quantification in Metallic Additive Manufacturing Through Physics-Informed Data-Driven Modeling [J].
Wang, Zhuo ;
Liu, Pengwei ;
Ji, Yanzhou ;
Mahadevan, Sankaran ;
Horstemeyer, Mark F. ;
Hu, Zhen ;
Chen, Lei ;
Chen, Long-Qing .
JOM, 2019, 71 (08) :2625-2634
[44]   Face Aging with Identity-Preserved Conditional Generative Adversarial Networks [J].
Wang, Zongwei ;
Tang, Xu ;
Luo, Weixin ;
Gao, Shenghua .
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :7939-7947
[45]  
Welsch G., 1993, MAT PROPERTIES HDB T
[46]  
Wohlers T. T., 2019, WOHLERS REP
[47]   Real-Time Monitoring of High-Dimensional Functional Data Streams via Spatio-Temporal Smooth Sparse Decomposition [J].
Yan, Hao ;
Paynabar, Kamran ;
Shi, Jianjun .
TECHNOMETRICS, 2018, 60 (02) :181-197
[48]   Anomaly Detection in Images With Smooth Background via Smooth-Sparse Decomposition [J].
Yan, Hao ;
Paynabar, Kamran ;
Shi, Jianjun .
TECHNOMETRICS, 2017, 59 (01) :102-114
[49]   Data-driven multi-scale multi-physics models to derive process-structure-property relationships for additive manufacturing [J].
Yan, Wentao ;
Lin, Stephen ;
Kafka, Orion L. ;
Lian, Yanping ;
Yu, Cheng ;
Liu, Zeliang ;
Yan, Jinhui ;
Wolff, Sarah ;
Wu, Hao ;
Ndip-Agbor, Ebot ;
Mozaffar, Mojtaba ;
Ehmann, Kornel ;
Cao, Jian ;
Wagner, Gregory J. ;
Liu, Wing Kam .
COMPUTATIONAL MECHANICS, 2018, 61 (05) :521-541
[50]   StackGAN plus plus : Realistic Image Synthesis with Stacked Generative Adversarial Networks [J].
Zhang, Han ;
Xu, Tao ;
Li, Hongsheng ;
Zhang, Shaoting ;
Wang, Xiaogang ;
Huang, Xiaolei ;
Metaxas, Dimitris N. .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2019, 41 (08) :1947-1962