About this Abstract |
Meeting |
2024 TMS Annual Meeting & Exhibition
|
Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
Presentation Title |
F-13: A Machine Learning Based Approach for Accelerated Textured Microstructure Generation |
Author(s) |
Gregory D. Wong, Anthony D. Rollett, Gregory S. Rohrer |
On-Site Speaker (Planned) |
Gregory D. Wong |
Abstract Scope |
Additive manufacturing is a strong example of the processing-structure-properties relationship at the core of materials science. Minor changes to process parameters can massively affect the microstructure and, consequently, the properties of parts, even while using parameters that create a fully dense part. Existing computational methodologies for predicting the process-microstructure relationship come at great computational expense. This leads to machine learning as a low-cost computational method to bridge the gap between process and microstructure. This work presents the use of GAN-type machine learning models to predict novel microstructures including crystallographic texture from sets of processing parameters. These machine learning generated microstructures are then compared to those obtained through analytical computational models with respect to their mechanical properties calculated using a Fourier transform-based micromechanical model. |
Proceedings Inclusion? |
Planned: |
Keywords |
Additive Manufacturing, Machine Learning, |