Deep learning model for predicting the spatial distribution of binding energy from atomic configurations
Abstract Understanding plasma-material interaction is crucial for achieving steady-state operation of magnetic confinement fusion devices. Kinetic Monte Carlo (kMC) simulation is a powerful tool for investigating the motion of atoms in the plasma facing materials under the influence of this interaction. To predict trapping sites and migration energies necessary for kMC simulations, we developed a deep learning model based on pix2pix for predicting the spatial distribution of binding energy. Results show that the model can reproduce spatial distributions similar to the true values. However, larger errors occur in regions with steep value gradients.
Authors
- K. Hoshino (ORCID: https://orcid.org/0000-0001-5672-9640)
- K. Sawada (ORCID: https://orcid.org/0000-0002-9469-6912)
- Masahiro Hasuo (ORCID: https://orcid.org/0000-0001-8441-2640)
- C. Takahashi (ORCID: https://orcid.org/0000-0003-3973-4856)
- Hiroaki Nakamura (ORCID: https://orcid.org/0000-0002-0593-8810)
- Masahiro Kobayashi (ORCID: https://orcid.org/0000-0002-6618-7387)
- Seiki Saitô
- Shingo Sato
Institutions
- Yamagata University (JP)
- Shinshu University (JP)
- Keio University (JP)
- Kyoto University (JP)
- National Institute for Fusion Science (JP)
- Ōtani University (JP)
- Yamagata University Hospital (JP)
- Nagoya University (JP)
Publication Details
- Journal
- Japanese Journal of Applied Physics
- Published
- 2024-08-13
- DOI
- https://doi.org/10.35848/1347-4065/ad6e8e
- Citations
- 1
- Primary Topic
- Machine Learning in Materials Science
- Type
- article
- Field-Weighted Citation Impact
- 0.12