About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Novel Strategies for Rapid Acquisition and Processing of Large Datasets From Advanced Characterization Techniques
|
| Presentation Title |
High-Throughput Time-Resolved X-Ray Computed Tomography to Characterize Flaw Evolution in LPBF Parts During Creep |
| Author(s) |
Rahul Franklin, Chase Joslin, Andres Marquez Rossy, Holden Hyer, Caleb Massey, Peeyush Nandwana, Alex Plotkowski, Sebastien Dryepondt, Amirkoushyar Ziabari |
| On-Site Speaker (Planned) |
Rahul Franklin |
| Abstract Scope |
Components printed using laser powder bed fusion (LPBF) must undergo a strict qualification process for use in high-temperature settings. X-ray computed tomography (XCT) is a powerful tool to characterize microstructures in 3D. However, the time needed for data acquisition and image quantification can be prohibitive for high-throughput industrial qualification processes. Here we present the workflows developed at ORNL to rapidly acquire and analyze time-resolved 3D data during interrupted creep tests of LPBF printed Haynes-282 and 316H SS parts. This consists of:
1) Fast XCT scans using an ORNL developed AI powered XCT reconstruction framework called Simurgh
2) Image registration to align micro-structural features in datasets to track flaw evolution
3) Image segmentation to classify the pores and material voxels
4) Image quantification to extract flaw attributes of interest
Through this workflow, individual flaws and overall creep deformation behavior can be accurately tracked and quantified between the different time-steps. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Characterization, Additive Manufacturing, Machine Learning |