[3O01-06] Cladding Materials 1

11:30 AM - 11:45 AM JST(2:30 AM - 2:45 AM UTC)

[3O06] Development of potential function for SiC materials used in nuclear applications based on machine learning

*Baopu WANG¹, Liangfan Zhu¹, Yuting Chen¹, Hirotomo Iwakiri², Kazunori Morishita¹ (1. Kyoto Univ., 2. Ryukyus Univ.)

Keywords:

SiC material,First-principle calculation,Machine learning,Potential function,Irradiation damage

SiC materials possess excellent properties in many respects and are being developed as nuclear materials. To study effects of nuclear environment on SiC, molecular dynamics method can analyze the behavior of radiation damage. However, the reliability of MD method depends on the accuracy of the interatomic potential functions used. First-principles calculations can provide high accuracy but require good computational speed and system size. Meanwhile, with the advancement of AI technology, machine learning methods can enhance both speed and scale. Therefore, this research combines first-principles calculations with machine learning, which is inputting data from first-principles calculations into a neural network for training and validation and developing a potential function, aims to apply it to MD simulations of radiation damage.