| About this Abstract |
| Meeting |
2023 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2023)
|
| Symposium
|
2023 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2023)
|
| Presentation Title |
Machine Learning Assisted Predictions of Porosity and Mechanical Properties Using In – Situ Photodiode |
| Author(s) |
Alex M. Butler, Brandon McWilliams |
| On-Site Speaker (Planned) |
Alex M. Butler |
| Abstract Scope |
Laser powder bed fusion (LPBF) can deliver optimized, complex geometries and is a topic of research programs for the Army. However, creating consistent and reliable parts at the point of need could require an experienced technician to monitor prints. Utilizing in-situ monitoring and mapping from in-situ data signal to part quality or properties, can allow for higher confidence in production. AlSi10Mg test coupon cylinders were printed while collecting layer wise in-situ data from a coaxial photodiode. Machine learning (ML) regression models will use print parameters and in-situ data to predict porosity in the samples. The best model will be down selected, and active learning implemented limiting the required labeled data to make accurate predictions of porosity. After this effort, dog bone samples will be printed, in-situ data collected, and a similar ML framework followed to map print parameters and layer wise in-situ data to mechanical properties. |
| Proceedings Inclusion? |
Definite: Post-meeting proceedings |