Physics-informed machine learning-based real-time long-horizon temperature fields prediction in metallic additive manufacturing
Guardat en:
| Publicat a: | Communications Engineering vol. 4, no. 1 (Dec 2025), p. 168 |
|---|---|
| Autor principal: | |
| Altres autors: | , , , , |
| Publicat: |
Springer Nature B.V.
|
| Matèries: | |
| Accés en línia: | Citation/Abstract Full Text Full Text - PDF |
| Etiquetes: |
Sense etiquetes, Sigues el primer a etiquetar aquest registre!
|
MARC
| LEADER | 00000nab a2200000uu 4500 | ||
|---|---|---|---|
| 001 | 3255943831 | ||
| 003 | UK-CbPIL | ||
| 022 | |a 2731-3395 | ||
| 024 | 7 | |a 10.1038/s44172-025-00501-7 |2 doi | |
| 035 | |a 3255943831 | ||
| 045 | 2 | |b d20251201 |b d20251231 | |
| 100 | 1 | |a Tian, Mingxuan |u Nanjing Tech University, School of Mechanical and Power Engineering, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210) | |
| 245 | 1 | |a Physics-informed machine learning-based real-time long-horizon temperature fields prediction in metallic additive manufacturing | |
| 260 | |b Springer Nature B.V. |c Dec 2025 | ||
| 513 | |a Journal Article | ||
| 520 | 3 | |a Real-time long-horizon temperature prediction in wire arc additive manufacturing is critical for process control and quality assurance. However, finite element methods are computationally expensive, and the existing data-driven models suffer from error accumulation and poor adaptability. Here we propose a physics-informed geometric recurrent neural network that integrates geometric characteristics and physical constraints, captures spatiotemporal characteristics via convolutional long short-term memory cells, and enforces physical consistency through hard-encoding initial/boundary conditions and physics-informed loss function. The model can predict the temperature field for future 1.25 s based on current 1.25 s data, and has also been evaluated for more long-horizon predictions. Transfer learning was used to enhance the model’s efficiency in practical applications. Results demonstrate that the proposed model achieves 4.5−13.9% maximum prediction error in simulations and experimental data. Including geometric characteristics and physical information reduces maximum error by about 1%, while the integrated model lowers it by 4%. Furthermore, transfer learning reduces the training time by approximately 50% while achieving the same loss level.Real-time long-horizon temperature prediction in metal additive manufacturing is critical for process control and quality assurance. Mingxuan Tian and colleagues propose a physics-informed machine learning model to predict temperature field for future 1.25 s. | |
| 653 | |a Temperature distribution | ||
| 653 | |a Finite element method | ||
| 653 | |a Finite volume method | ||
| 653 | |a Accuracy | ||
| 653 | |a Fluid dynamics | ||
| 653 | |a Parameter identification | ||
| 653 | |a Quality assurance | ||
| 653 | |a Boundary conditions | ||
| 653 | |a Data processing | ||
| 653 | |a Manufacturing | ||
| 653 | |a Machine learning | ||
| 653 | |a Low carbon steel | ||
| 653 | |a Additive manufacturing | ||
| 653 | |a Efficiency | ||
| 653 | |a Simulation | ||
| 653 | |a Physics | ||
| 653 | |a Partial differential equations | ||
| 653 | |a Residual stress | ||
| 653 | |a Temperature | ||
| 653 | |a Neural networks | ||
| 653 | |a Process controls | ||
| 653 | |a Recurrent neural networks | ||
| 653 | |a Data collection | ||
| 653 | |a Finite element analysis | ||
| 653 | |a Errors | ||
| 653 | |a Real time | ||
| 700 | 1 | |a Mu, Haochen |u Nanjing Tech University, School of Mechanical and Power Engineering, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210); Nanjing Tech University, Institute of Reliability centered Manufacturing, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210); Nanjing Tech University, Jiangsu Provincial Key Laboratory of Energy Power Manufacturing Equipment and Reliability Technology, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210) | |
| 700 | 1 | |a Liu, Tao |u Nanjing Tech University, School of Mechanical and Power Engineering, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210) | |
| 700 | 1 | |a Li, Mengjiao |u Nanjing Tech University, School of Mechanical and Power Engineering, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210) | |
| 700 | 1 | |a Ding, Donghong |u Nanjing Tech University, School of Mechanical and Power Engineering, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210); Nanjing Tech University, Institute of Reliability centered Manufacturing, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210); Nanjing Tech University, Jiangsu Provincial Key Laboratory of Energy Power Manufacturing Equipment and Reliability Technology, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210) | |
| 700 | 1 | |a Zhao, Jianping |u Nanjing Tech University, School of Mechanical and Power Engineering, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210); Nanjing Tech University, Jiangsu Provincial Key Laboratory of Energy Power Manufacturing Equipment and Reliability Technology, Nanjing, China (GRID:grid.412022.7) (ISNI:0000 0000 9389 5210) | |
| 773 | 0 | |t Communications Engineering |g vol. 4, no. 1 (Dec 2025), p. 168 | |
| 786 | 0 | |d ProQuest |t Science Database | |
| 856 | 4 | 1 | |3 Citation/Abstract |u https://www.proquest.com/docview/3255943831/abstract/embedded/6A8EOT78XXH2IG52?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text |u https://www.proquest.com/docview/3255943831/fulltext/embedded/6A8EOT78XXH2IG52?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text - PDF |u https://www.proquest.com/docview/3255943831/fulltextPDF/embedded/6A8EOT78XXH2IG52?source=fedsrch |