About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
Presentation Title |
Application of Multi-Physics Simulations and Machine Learning to Predict Spatter in Laser Powder Bed Fusion |
Author(s) |
Olabode Ajenifujah, Francis Ogoke, Jack Beuth, Amir Barati-Farimani |
On-Site Speaker (Planned) |
Olabode Ajenifujah |
Abstract Scope |
Laser powder bed fusion (LPBF) is the predominant metal additive manufacturing (AM) method. Nevertheless, its widespread adoption for manufacturing parts across various sectors is restricted by defects that impair the mechanical properties of the components. Spatters occur due to the complex melt pool dynamics during laser-material interactions in LPBF. Spatter generation is known to promote the formation of defects such as porosity and roughness. In this presentation, we will detail the analysis of spatter and melt pool using datasets obtained from modeling LPBF processes with an open-source tool, OpenFOAM. Then, we will discuss the prediction of spatter properties coupled with the mechanistic understanding gained from applying our dataset to regression and classification tasks using machine learning models. Our research concludes with the creation of a detailed surrogate model of spatter and a process map to adjust processing parameters and reduce defects, thus improving AM's practicality for industrial uses. |
Proceedings Inclusion? |
Planned: |