| About this Abstract |
| Meeting |
MS&T24: Materials Science & Technology
|
| Symposium
|
Additive Manufacturing: Equipment, Instrumentation and In-Situ Process Monitoring
|
| Presentation Title |
Semantic Segmentation and Spreading Anomaly Identification of Binder Jet In-Situ Images |
| Author(s) |
Alexander Gourley, Jonathan Kaufman, Bashu Aman, Edwin Schwalbach, Jack Beuth, Lisa Rueschhoff, Reeja Jayan |
| On-Site Speaker (Planned) |
Alexander Gourley |
| Abstract Scope |
Variability in the inherently dynamic nature of additive manufacturing introduces imperfections that hinder the commercialization of new materials. Binder jetting has produced ceramic and metallic parts, but low green densities and spreading anomalies reduce the predictability and processability of resulting geometries. In-situ feedback presents a method for robust evaluation of spreading anomalies, reducing the required builds to refine processing parameters in a multivariate space. In this study, we generated and compared U-Net semantic segmentation models for visually different powders using single builds for training data, identifying the challenges of extending existing segmentation methods to visually lighter
oxide powders. Leveraging preexisting analysis tools allowed for rapid analysis of oxide powder by providing an accessible framework for implementing neural networks. Robust analysis techniques and the demonstration of correcting anomalies with processing parameters show promise for developing automation with in-situ feedback. |