| About this Abstract |
| Meeting |
2024 TMS Annual Meeting & Exhibition
|
| Symposium
|
2024 Technical Division Student Poster Contest
|
| Presentation Title |
SPG-80: Utilizing Machine Learning Techniques to Correlate Constituent Redistribution, Fission Gas Bubble Structures, and Thermal Conductivity Changes in Annular, Irradiated U-Zr Fuels |
| Author(s) |
Mary Elizabeth Sevart, Mitch Mika, Fei Xu, Tiankai Yao, Luca Capriotti, Assel Aitkaliyeva |
| On-Site Speaker (Planned) |
Mary Elizabeth Sevart |
| Abstract Scope |
Uranium-zirconium (U-Zr) fuels may be considered for use in both open and closed fuel cycles. In a closed fuel cycle, solid U-Zr fuels have a relatively large gap between fuel and cladding filled with sodium to facilitate heat transport. An open fuel cycle relies on an as-built gap in the center of the fuel filled with helium rather than sodium. Preliminary data has shown that the constituent redistribution and microstructure evolution in an annular geometry may be different than its solid counterpart. In solid fuel designs, three constituent redistribution zones appear due to the radial temperature gradient driving zirconium redistribution as compared to two primary zones in annular fuels. A machine learning algorithm is presented here to quantify the size, number density, and structure of fission gas bubbles in each of the constituent redistribution zones and connect this to local thermal conductivity degradation in three different annular, irradiated U-Zr fuels. |
| Proceedings Inclusion? |
Undecided |
| Keywords |
Nuclear Materials, Characterization, Machine Learning |