About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
Presentation Title |
Effect of nucleation model and data resolution on cellular automata texture strength prediction |
Author(s) |
Matthew Rolchigo, John Coleman, Gerald Knapp, Alex Plotkowski |
On-Site Speaker (Planned) |
Matthew Rolchigo |
Abstract Scope |
Cellular automata (CA) models for predicting grain structure in additively manufactured parts have become widely used and validated against observed grain structures for many alloys and processing conditions. However, for some conditions, particularly for those with complex scan paths and scan path rotation between layers, CA models often underpredict the texture observed from experiments. This study investigates potential causes of and solutions to this discrepancy, including underresolution of the undercooled liquid at coarse cell sizes, inaccuracy in approximating dendrite branching within grain envelopes, and inaccuracy in the nucleation and impingement behavior at the fusion line boundaries. Results of a cell size sensitivity study will be presented along with example simulation results using alternative approaches to nucleation modeling performed using the ExaCA software. Microstructure predictions will be compared to data collected at ORNL’s Manufacturing Demonstration Facility, and a path for calibrating CA inputs will be discussed. Funded by US DOE EERE. |
Proceedings Inclusion? |
Planned: |
Keywords |
Additive Manufacturing, Computational Materials Science & Engineering, Solidification |