About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing Modeling, Simulation and Artificial Intelligence
|
| Presentation Title |
Identifying LPBF Anomalies and Creep Damage in Nickel Based Super Alloys |
| Author(s) |
Amanda Heimbrook, Holden T Hyer, Rahul T Franklin, Amir T Ziabari, Sebastien T Dryepondt |
| On-Site Speaker (Planned) |
Amanda Heimbrook |
| Abstract Scope |
To support the qualification of materials for advanced nuclear reactor concepts, γ’ strengthened Haynes 282 and solid solution strengthened Inconel 625 are of interest due to their superior high temperature strength. Haynes 282 parts positioned closest to the argon flow inlet during the LPBF process had fewer flaws compared to parts located further away due to spatter. Those printing defects were only exacerbated during creep testing. In contrast, LPBF Inconel 625 was nearly defect-free across the entire build plate with creep properties similar to its wrought counterpart. Image analysis was used to detect anomalies from optical images. A machine learning model was trained to distinguish between printing anomalies versus creep damage (cracks and cavitation) after testing. In parallel, X-ray computed tomography was used to capture larger printing defects, as well as the coalescence of smaller printed flaws, to help predict where failure would occur during an interrupted creep test. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, Machine Learning, Nuclear Materials |