| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
| Presentation Title |
Melt Pool Width Prediction With Machine Learning In Selective Laser Melting |
| Author(s) |
Umut Can Gulletutan, Sertaç Altınok, Irmak Sargın |
| On-Site Speaker (Planned) |
Umut Can Gulletutan |
| Abstract Scope |
Precise control of melt pool dynamics in Selective Laser Melting (SLM) is critical for achieving high-quality final parts, particularly when shaping intricate geometries. This necessitates accurate prediction of melt pools. To address this challenge, a dataset of 1266 points, including melt pool width, process parameters, and material properties, was compiled from scientific literature for various alloy compositions. The traditional Rosenthal equation while offering an easy analytical solution, suffers from lower prediction accuracy, R2 = 0.63 due to assumptions of steady-state conditions and homogeneous material properties. Conversely, a Random Forest machine learning model trained with the dataset achieved superior accuracy with an R2 of 0.92. Experimental validation further corroborated the limitations of the Rosenthal equation compared to machine learning. Overall, the machine learning approach significantly enhanced melt pool width predictions, emphasizing the importance of advanced techniques for optimizing SLM processes to achieve higher precision in additive manufacturing. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, Machine Learning, Characterization |