| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
| Presentation Title |
AiDED: Accurate Machine Learning Inference Framework for Process Parameter Optimization in Laser Directed Energy Deposition |
| Author(s) |
Xiao Shang, Evelyn Li, Ajay Talbot, Haitao Wen, Tianyi Lyu, Jiahui Zhang, Yu Zou |
| On-Site Speaker (Planned) |
Xiao Shang |
| Abstract Scope |
In additive manufacturing, process optimization is a long-standing challenge. Conventional process optimization approaches require significant numbers of experiments or computations, and often lack the versatility to address multiple optimization objectives by production needs. Although efforts have been made using machine learning methods, most work to date only focuses on single-track prints, which hardly translates to multi-track and multi-layer cases in real-life applications. Here, we propose an accurate process optimization framework based on machine learning inference models (AiDED), to aide the process parameter optimization in laser directed energy deposition (L-DED). With AiDED, one can: 1. Accurately predict single-track, multi-track, and multi-layer geometries directly from process parameters; 2. Determine hatch-spacing and layer thickness for fabricating fully dense multi-track and multi-layer prints; 3. Identify optimal process parameters based on customizable optimization objectives. With the transferability of AiDED, we hope that anyone can easily develop their own version of AiDED to “aide” their applications. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, Machine Learning, |