| About this Abstract |
| Meeting |
2024 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
| Presentation Title |
Application of Computer Vision to Mapping of Process Parameters to Material Structure of AM Carbon Fiber Composites |
| Author(s) |
Kenneth M. Clarke, Michael Groeber, John Wertz , Michael Chapman , Andrew Abbott , Roneisha Haney |
| On-Site Speaker (Planned) |
Kenneth M. Clarke |
| Abstract Scope |
Machine learning continues to find applications in exploration of the process structure property linkages of materials. Additive manufacturing (AM) has recently gained the attention of industries and researchers as a useful alternative to conventional manufacturing. However, the effects of AM process parameters on structure still remain under-researched . This research project applies machine learning models to experimental images from a direct ink writing process of a carbon fiber composite epoxy. Since machine learning performs well at pattern recognition tasks, the belief is the same models could be applied to better understand the process to structure linkage. The meso-scale of the material is the main focus of this research as this scale captures higher-order features of descriptors such as clustering and texturing. The workflow encodes bulk statistics of the microstructure into the meso-scale representations through the use of RGB images to facilitate a linkage between the varying length scales. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, Polymers, Machine Learning |