[3N07-09] Data Science and Fuel Properties

11:25 AM - 11:40 AM JST(2:25 AM - 2:40 AM UTC)

[3N08] New Developments in Nuclear Fuel Research through Integration with Data Science(11)Selection of novel high thermal conductivity nuclear fuels based on machine learning and physical property evaluation of U2Ti

*Ryusuke Torata¹, Yuji Ohishi¹, Hiroaki Muta¹, Yifan Sun², Masaya Kumagai^2,3,4, Ken Kurosaki² (1. Osaka University Graduate School of Engineering, 2. Kyoto University Institute for Integrated Radiation and Nuclear Science, 3. Sakura Internet, 4. RIKEN)

Keywords:

machine learning,nuclear fuel materials,uranium compounds,U2Ti,high thermal conductivity

U₂Ti was selected from uranium compounds predicted to have high thermal conductivity by machine learning. High-density single-phase samples were prepared using arc melting and spark plasma sintering. The temperature dependence of the thermal conductivity was evaluated using the laser flash method and compared with the machine learning predictions.