About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing Modeling, Simulation and Artificial Intelligence
|
| Presentation Title |
Surrogate Modeling for Computationally Efficient Prediction of Thermal History and Molten Pool Shape in a Large Domain |
| Author(s) |
Corbin Grohol, Yung C. Shin |
| On-Site Speaker (Planned) |
Yung C. Shin |
| Abstract Scope |
This study covers predicting accurate temperature fields in a large domain using a surrogate modeling technique for metal laser powderbed fusion processes. Simulation of large-scale components via high-fidelity modeling involves prohibitive high computing costs even with massively parallelized computing. Computationally efficient low-fidelity models, however, cannot predict temperature history and molten pool shapes accurately. To overcome this challenge, a surrogate modeling approach is developed using a lower-fidelity model to extract temperature variation in a large domain with pertinent features and implement an active learning algorithm to determine when and where the high-fidelity model needs to be simulated to improve modeling results. Using such an approach, the computational load is decreased significantly, increasing calculation throughput. With this approach, location and time dependent molten pool shapes over a large domain are accurately predicted with affordable computational time. The validation results are provided to show that this surrogate method is effective and accurate. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, Machine Learning, Modeling and Simulation |