| About this Abstract |
| Meeting |
TMS Specialty Congress 2025
|
| Symposium
|
Joint Sessions of AIM, ICME, & 3DMS
|
| Presentation Title |
Harnessing Multi-Modal Metrology Data for Predictive Modeling in Laser Powder Directed Energy Deposition |
| Author(s) |
Michael Juhasz, Eric Chin, Youngsoo Choi, Joseph T. McKeown, Saad Khairallah |
| On-Site Speaker (Planned) |
Michael Juhasz |
| Abstract Scope |
In this presentation, we utilize extensive multi-modal on-machine metrology data from Laser Powder Directed Energy Deposition (LP-DED) to create a robust surrogate model for the 3D printing process. By employing Dynamic Mode Decomposition with Control (DMDc), we capture the intricate physics within this dataset. Our physics-based model focuses on key thermodynamic parameters, enabling accurate predictions of critical outcomes. It integrates 21 process parameters, including laser power and scan rate, to predict outputs like melt pool temperature and size. The model also incorporates uncertainty quantification, providing reliable bounds on predictions and enhancing confidence in results. We validate the model by applying it to a new part and monitoring the printing process, finding that its predictions closely align with actual measurements. Prepared by LLNL under Contract DE-AC52-07NA27344. |
| Proceedings Inclusion? |
Definite: Post-meeting proceedings |