Three-Dimensional Modeling and Analysis of Directed Energy Deposition Melt Pools Based on Physical Information Neural Networks
Source abstract
Abstract
In Directed Energy Deposition (DED), modeling the molten pool temperature field is crucial for precise temperature control, process optimization, and quality improvement. However, conventional numerical methods suffer from limitations such as high computational costs and poor transferability. This study proposes a physics-informed neural network with dynamic learning rate (DLR-PINN) model, which integrates transfer learning to enable rapid prediction of 3D temperature fields and dimensions of molten pools across process parameters. Its validity is verified by a finite element method (FEM) calibrated via single-track DED experiments. Results show that DLR-PINN exhibits superior convergence and stability compared to traditional PINN. Combined with transfer learning, training efficiency is significantly enhanced, with a single prediction taking only 10 s. Using the FEM as the benchmark, it achieves a mean absolute percentage error (MAPE) of 0.53% for temperature prediction, and MAPE of 3.69%, 2.48%, and 6.96% for molten pool dimension predictions, respectively. Sensitivity analysis of process parameters reveals that scanning speed has a significantly greater regulatory effect on molten pool characteristics than laser power. Additionally, the temperature field of the flat-top heat source is more uniform than that of the Gaussian heat source, which is more conducive to improving printing quality and efficiency.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Additive Manufacturing Materials and Processes · Advancements in Semiconductor Devices and Circuit Design · Advancements in Photolithography Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
State Key Laboratory of Featured Metal Materials and Life-Cycle Safety for Composite Structures, Guangxi University, Nanning 530004, China
Institute of Laser Intelligent Manufacturing and Precision Processing, School of Mechanical Engineering, Guangxi University, Nanning 530004, China
Funding
National Key R&D Program of China · Key R&D Program of Guangxi Province · Internet Information Innovation Talent of Guangxi Province · Guangxi Natural Science Foundation
References (50)
Show source-provided references
1. Beam Shaping Technology and Its Application in Metal Laser Additive Manufacturing: A Review10.1016/j.jmrt.2023.08.037 ↗
2. Directed Energy Deposition (DED) Additive Manufacturing: Physical Characteristics, Defects, Challenges and Applications10.1016/j.mattod.2021.03.020 ↗
3. Formation Mechanism of Inherent Spatial Heterogeneity of Microstructure and Mechanical Properties of NiTi SMA Prepared by Laser Directed Energy Deposition10.1088/2631-7990/acd96f ↗
4. Review on Residual Stresses in Metal Additive Manufacturing: Formation Mechanisms, Parameter Dependencies, Prediction and Control Approaches10.1016/j.jmrt.2022.02.054 ↗
5. Kiran, A., Hodek, J., Vavřík, J., Urbánek, M., and Džugan, J. (2020). Numerical Simulation Development and Computational Optimization for Directed Energy Deposition Additive Manufacturing Process. Materials, 13.10.3390/ma13112666 ↗
6. Shu, Y., Galles, D., Tertuliano, O.A., McWilliams, B.A., Yang, N., Cai, W., and Lew, A.J. (2021). A Critical Look at the Prediction of the Temperature Field around a Laser-Induced Melt Pool on Metallic Substrates. Sci. Rep., 11.10.1038/s41598-021-91039-z ↗
7. Evolution of Melt Pool Motion and Temperature Field Based on Powder Scale Modeling during Laser Directed Energy Deposition Process10.1016/j.applthermaleng.2024.122564 ↗
8. High-Fidelity Modelling of Thermal Stress for Additive Manufacturing by Linking Thermal-Fluid and Mechanical Models10.1016/j.matdes.2020.109185 ↗
9. Research and Application of Machine Learning for Additive Manufacturing
10. Machine Learning Algorithms for Defect Detection in Metal Laser-Based Additive Manufacturing: A Review10.1016/j.jmapro.2021.12.061 ↗
11. Machine Learning in Additive Manufacturing: State-of-the-Art and Perspectives
12. Thermal Field Prediction for Laser Scanning Paths in Laser Aided Additive Manufacturing by Physics-Based Machine Learning10.1016/j.cma.2019.112734 ↗
13. Prediction of Melt Pool Temperature in Directed Energy Deposition Using Machine Learning
14. Data-Driven Prognostic Model for Temperature Field in Additive Manufacturing Based on the High-Fidelity Thermal-Fluid Flow Simulation10.1016/j.cma.2022.114652 ↗
15. Machine Learning for Metal Additive Manufacturing: Towards a Physics-Informed Data-Driven Paradigm10.1016/j.jmsy.2021.11.003 ↗
16. Role of Machine Learning in Additive Manufacturing of Titanium Alloys—A Review10.1007/s11831-023-09969-y ↗
17. Physics-Informed Machine Learning10.1038/s42254-021-00314-5 ↗
18. Meng, C., Seo, S., Cao, D., Griesemer, S., and Liu, Y. (2022). When Physics Meets Machine Learning: A Survey of Physics-Informed Machine Learning. arXiv.
19. Hao, Z., Liu, S., Zhang, Y., Ying, C., Feng, Y., Su, H., and Zhu, J. (2023). Physics-Informed Machine Learning: A Survey on Problems, Methods and Applications. Methods Appl.
20. Physics-Informed Neural Networks: A Deep Learning Framework for Solving Forward and Inverse Problems Involving Nonlinear Partial Differential Equations10.1016/j.jcp.2018.10.045 ↗
21. Simulation of Multi-Species Flow and Heat Transfer Using Physics-Informed Neural Networks10.1063/5.0058529 ↗
22. Physics-Informed Neural Networks (PINNs) for Fluid Mechanics: A Review10.1007/s10409-021-01148-1 ↗
23. Physics-Informed Neural Networks for Phase-Field Method in Two-Phase Flow10.1063/5.0091063 ↗
24. Physics-Informed Neural Networks for Heat Transfer Problems10.1115/1.4050542 ↗
25. Misyris, G.S., Venzke, A., and Chatzivasileiadis, S. (2020, January 2–6). Physics-Informed Neural Networks for Power Systems. Proceedings of the 2020 IEEE Power & Energy Society General Meeting (PESGM), Montreal, QC, Canada.10.1109/PESGM41954.2020.9282004 ↗
26. Machine Learning for Metal Additive Manufacturing: Predicting Temperature and Melt Pool Fluid Dynamics Using Physics-Informed Neural Networks10.1007/s00466-020-01952-9 ↗
27. 3D Temperature Field Prediction in Direct Energy Deposition of Metals Using Physics Informed Neural Network10.1007/s00170-021-08542-w ↗
28. Hybrid Thermal Modeling of Additive Manufacturing Processes Using Physics-Informed Neural Networks for Temperature Prediction and Parameter Identification10.1007/s00466-022-02257-9 ↗
29. A Physics-Informed Neural Network Framework to Predict 3D Temperature Field without Labeled Data in Process of Laser Metal Deposition10.1016/j.engappai.2023.105908 ↗
30. Effect of Stress Relaxation on Distortion in Additive Manufacturing Process Modeling
31. In Situ Measurements and Thermo-Mechanical Simulation of Ti–6Al–4V Laser Solid Forming Processes10.1016/j.ijmecsci.2019.01.043 ↗
32. Microstructure Prediction of Selective Laser Melting AlSi10Mg Using Finite Element Analysis10.1016/j.matdes.2018.01.022 ↗
33. Residual Stress Reduction and Surface Quality Improvement of Dual-Laser Powder Bed fusion
34. Prediction of Residual Stress and Deformation Based on the Temperature Distribution in 3D-Printed Parts10.1007/s00170-021-06711-5 ↗
35. PyTorch Core Team (2025, August 12). torch.optim.lr_scheduler. ReduceLROnPlateau. PyTorch Documentation. Available online: https://pytorch.org/docs/stable/generated/torch.optim.lr_scheduler.ReduceLROnPlateau.html.
36. Laser Additive Manufacturing of Titanium Alloys: Process, Materials and Post-Processing10.1007/s12598-024-02685-x ↗
37. Operando Synchrotron X-Ray Analysis of Melt Pool Dynamics in an al-Sn Immiscible Alloy
38. Investigation of Thermal and Fluid Dynamics Behaviors of Melt Pool during Laser Direct Metal Deposition on Inclined Substrates10.1016/j.mtcomm.2024.110953 ↗
39. Simulation of Melt Pool Dynamics Including Vaporization Using the Particle Finite Element Method
40. Laser Remelting of CoCrFeNiTi High Entropy Alloy Coatings Fabricated by Directed Energy Deposition: Effects of Remelting Laser Power10.1016/j.optlastec.2022.108871 ↗
41. Review on Laser Directed Energy Deposited Aluminum Alloys10.1088/2631-7990/ad16bb ↗
42. Saboori, A., Aversa, A., Marchese, G., Biamino, S., Lombardi, M., and Fino, P. (2020). Microstructure and Mechanical Properties of AISI 316L Produced by Directed Energy Deposition-Based Additive Manufacturing: A Review. Appl. Sci., 10.10.3390/app10093310 ↗
43. Dynamics of Molten Pool Evolution on Flat Plates and Outer/Inner Shaft Surfaces in High-Speed Laser Cladding
44. Effects of Laser Scanning Speeds on Different States of the Molten Pool during Selective Laser Melting: Simulation and Experiment10.1016/j.matdes.2020.108542 ↗
45. Effect of Beam Energy Density Characteristics on Microstructure and Mechanical Properties of Nickel-Based Alloys Manufactured by Laser Directed Energy Deposition10.1016/j.jmatprotec.2023.118074 ↗
46. Influence of Beam Diameter on Laser Powder Bed Fusion (L-PBF) Process
47. Mesoscopic Segregation in H13 Steel Molten Pool during Laser Remelting: A Combined Influence of Marangoni Convection and Oxidation10.1016/j.jmatprotec.2023.117956 ↗
48. A Framework to Link Localized Cooling and Properties of Directed Energy Deposition (DED)-Processed Ti-6Al-4V10.1016/j.actamat.2017.04.027 ↗
49. Stability of Molten Pool and Microstructure Evolution of Ti-6Al-4 V during Laser Powder Bed Fusion with a Flat-Top Beam
50. On Productivity of Laser Additive Manufacturing10.1016/j.jmatprotec.2018.05.033 ↗