Researchers of the Idaho National Laboratory (INL) and the University of Idaho successfully used machine learning to characterize the microstructure of metallic nuclear fuel, whose fine details are visible only under powerful magnification. The researchers chose uranium-zirconium fuel for the study.
The researchers developed machine learning approaches to extract and analyze a wide range of data points, such as the size and connectivity of fission gas bubbles, from irradiated uranium-zirconium fuel. Gas bubbles are a natural byproduct of nuclear fission along with smaller atoms including xenon and krypton. These and other byproducts are stored as bubbles within the fuel elements, resulting in microstructural changes that can limit the fuel ability to transfer heat.
To develop the machine learning process, the researchers first created a comprehensive dataset with high-resolution images of fuel cross-sections and manual annotations of fission gas bubbles. They then implemented a machine learning algorithm, known as decision tree, to predict the category of each bubble using quantitative image features including bubble size, shape, and appearance.
This study provides accurate information on the fuel morphology, fission gas bubble density, and lanthanide distribution. Lanthanides are elements that form during reactor operations and inhibit fuel efficiency. The developed models can be extended to perform accurate predictions of material properties. Characterizing the microstructure of irradiated fuel, particularly the bubbles and fission products that may form in operation, is crucial for engineering nuclear fuel that is both highly efficient and effectively resists fracture in potential accident conditions.
The project produced notable insights into uranium-zirconium fuel performance, including degradation of fuel thermal conductivity due to extensive pore structure and increase in connected pores in hotter fuel regions.
According to the Idaho National Laboratory