About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
2026 Technical Division Student Poster Contest
|
| Presentation Title |
SPU-23: Computer Vision of Virgin vs. Recovered Powder Particles |
| Author(s) |
Aileen Cornwall-Brady, Kyle L. Tsaknopoulos, Danielle L. Cote, Yasmin Soares |
| On-Site Speaker (Planned) |
Aileen Cornwall-Brady |
| Abstract Scope |
Cold spray additive manufacturing can be used to produce and restore metal parts by depositing atomized metallic powder at high velocity and relatively low temperatures. During this process, some particles effectively deposit on the substrate, while others rebound. While progress is being made on the collection of these unbonded particles, there is insufficient knowledge of the morphological differences from virgin powder. These differences could aid in determining effective treatments for recycling rebounded powder, leading to a more sustainable manufacturing process. To quantify morphological differences between virgin and rebounded powder particles, SEM images will be obtained in multiple locations. The images will be used with a computer vision model to segment individual powder particles and calculate morphological characteristics between the two powders. Traditional powder morphology characterization methods, including laser diffraction, dynamic image analysis, and BET analysis, will also be used and compared to data produced by the computer vision system. |
| Proceedings Inclusion? |
Undecided |
| Keywords |
Powder Materials, Characterization, Computational Materials Science & Engineering |