About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing: Advanced Characterization With Synchrotron, Neutron, and In Situ Laboratory-scale Techniques IV
|
| Presentation Title |
Real-Time Detection and Analysis of Spatter-Induced Defects in Laser-Based Powder Bed Fusion Using Optical and Long Wave Infrared Imaging and Machine Learning |
| Author(s) |
Cole Britt |
| On-Site Speaker (Planned) |
Cole Britt |
| Abstract Scope |
Laser-based powder bed fusion (PBF-LB) additive manufacturing (AM) produces stochastic material defects which can result in variability in mechanical properties. This work demonstrates the capabilities of in situ process monitoring using optical and infrared cameras for PBF. A machine learning method was trained on a variety of processing parameters and part geometries to detect and analyze various kinds of defects. A novel computer vision technique is also presented to track individual spatter particles ejected from the laser melt pool using long wave infrared (LWIR) imaging. This technique enables the examination of the effects of spatter particles on internal defects and their impacts on material properties. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, Machine Learning, Other |