| About this Abstract |
| Meeting |
2024 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
| Presentation Title |
Machine Learning-based Prediction of Evolution of Thermal Profiles During Additive Manufacturing |
| Author(s) |
Mani Krishna Karri, Aishwarya Manjunath, Shashank Sharma, Narendra B Dahotre |
| On-Site Speaker (Planned) |
Aishwarya Manjunath |
| Abstract Scope |
Laser Powder Bed Fusion (LPBF) is an additive manufacturing (AM) technology that depends on controlled temperature for optimal printed part properties and defect prevention. Accurate prediction of thermal profiles enables process optimization. Typically this is achieved through complex and time-consuming physics based finite element models. Application of machine learning (ML) for the prediction of these thermal profiles can speed up prediction capabilities but still remains largely unexplored. In this study, this is addressed through two approaches; first is through image translation, which predicts temperature distribution images from images with grid layouts, where each grid corresponds to a process parameter such as power, velocity etc. The second is through ML algorithm that can convert numerical values of process parameters into images of temperature distribution through up-sampling networks. Application of these methods is demonstrated in case of LPBF process modeling. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, ICME, Computational Materials Science & Engineering |