| About this Abstract |
| Meeting |
2024 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2024)
|
| Symposium
|
2024 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2024)
|
| Presentation Title |
Predicting Melt Pool Thermal Distribution in Ti-6Al-4V Directed Energy Deposition Using Machine Learning |
| Author(s) |
Sung-Heng Wu, Usman Tariq, Ranjit Joy, Muhammad Arif Mahmood, Frank Liou |
| On-Site Speaker (Planned) |
Sung-Heng Wu |
| Abstract Scope |
Ti-6Al-4V's high strength-to-weight ratio and thermophysical properties make it a compelling target for directed energy deposition (DED) components. The printed material's strength and fatigue properties are strongly related to the solidification and microstructure, which results from thermal gradients inherent to DED processes. Accurately predicting the melt pool thermal distribution and geometry is therefore crucial for enhancing the quality of the DED-fabricated parts. This study introduces the use of advanced machine learning models, including XGBoost, Gated Recurrent Units (GRU), Long Short-Term Memory (LSTM), and Bidirectional LSTM (Bi-LSTM), to predict and analyze the thermal distribution of the melt pool during the DED process with Ti-6Al-4V. The performance of each model was evaluated based on its computational efficiency and predictive accuracy. The comparative analysis of computational time and accuracy among the models provides insights into their practical applications, guiding future research and industrial applications in additive manufacturing. |
| Proceedings Inclusion? |
Definite: Post-meeting proceedings |