| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Microstructural, Mechanical, and Chemical Behavior of Solid Nuclear Fuel and Fuel-Cladding Interface II
|
| Presentation Title |
Bayesian Calibration of Ab-Initio Based Self- and Fission Gas Diffusivity in UN Nuclear Fuel |
| Author(s) |
Anton J. Schneider, Pieterjan Robbe, David Andersson, Christopher Matthews, Michael Cooper |
| On-Site Speaker (Planned) |
Anton J. Schneider |
| Abstract Scope |
Uranium mononitride (UN) is one of the ceramic nuclear fuel alternatives to oxide fuels considered for light water reactor and advanced reactor designs. Self- and fission gas diffusivity need to be quantitatively determined, given that they influence key fuel performance phenomena like swelling and fission gas release. Our recent study predicted diffusivities in UN, using ab-initio-informed cluster dynamics simulations. Yet, it is critical from a regulation standpoint to quantify the uncertainty of the underlying ab-initio results to attribute reliable error-bars to the derived engineering scale properties.
In this work, we train a surrogate model on hundreds of thousands of cluster dynamics simulations, and we carefully analyze the surrogate accuracy. The model is then used for the Bayesian calibration of the atomic scale parameters against available experimental literature, which leads not only to the uncertainty quantification of the diffusivities, but also to the likelihood distribution of every atomic scale parameter. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Modeling and Simulation, Nuclear Materials, |