| About this Abstract |
| Meeting |
2020 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing: Processing Effects on Microstructure and Material Performance
|
| Presentation Title |
A-99: Effect of Temperature Dependent Properties on the Accuracy of Physics-based Surrogate Models for Powder Bed Fusion Additive Manufacturing |
| Author(s) |
Alexander Wolfer, Richard Otis, Brian Weston, Saad Khairallah, Andy Anderson, Andrew A Shapiro, Jean-Pierre Delplanque |
| On-Site Speaker (Planned) |
Alexander Wolfer |
| Abstract Scope |
Low-order or surrogate models are often used to efficiently investigate various processing strategies for powder bed fusion processes, as well as to perform sensitivity analysis or uncertainty quantification. A common simplification used in many physics-based lower order models is to assume constant thermophysical properties, ignoring any temperature dependence. It is possible to capture such temperature dependency with high-fidelity simulations but at a prohibitive computational for applications that require a large number of cases to be considered (e.g. uncertainty quantification). An investigation into several approaches to account for the temperature dependence of thermophysical properties in a surrogate model is presented. We discuss the advantages and disadvantages of various modeling assumptions and how they affect the accuracy of the predictions (e.g. melt pool geometry or thermal gradients). Comparisons are made with high-fidelity simulations that include temperature dependent thermal properties as well as additional physics, such as phase change and melt-pool fluid dynamics. |
| Proceedings Inclusion? |
Planned: Supplemental Proceedings volume |