About this Abstract |
| Meeting |
MS&T25: Materials Science & Technology
|
| Symposium
|
Integrated Computational Materials Engineering for Physics-Based Machine Learning Models
|
| Presentation Title |
Accelerated Nuclear Materials Thermochemistry in MOOSE through Surrogate Modeling |
| Author(s) |
Parikshit Bajpai, Andrew Kitterman, Chaitanya Bhave, Daniel Schwen |
| On-Site Speaker (Planned) |
Parikshit Bajpai |
| Abstract Scope |
Understanding thermodynamic properties and resulting phase equilibria is crucial for simulating material microstructure, property evolution, and behavior, particularly in nuclear materials and reactors. Thermodynamic and kinetic information from CALPHAD databases has been used to inform various process and material models, with the coupling of CALPHAD calculations to multiphysics simulation tools such as the Multiphysics Object Oriented Simulation Environment (MOOSE) becoming increasingly significant in nuclear applications. However, due to the high cost of direct coupling of CALPHAD-based Gibbs energy minimization, the applications of this approach have been restricted to relatively small systems. To accelerate the coupling of CALPHAD calculations with MOOSE-based codes, a thermochemistry surrogate modeling framework is being developed. This work will demonstrate an on-the-fly surrogate modeling capability developed for the MOOSE framework and explore its applications to phase field and engineering scale simulations specifically focused on nuclear materials and reactors, enhancing the predictive capabilities vital for the nuclear industry. |