In Fusion, we are working to develop surrogate AI models to speed up traditional numerics for physical and technical research and applications.
The aim of the AI in Fusion research area is to develop surrogate AI models to speed up traditional numerics for physical and technical research and applications.
While numerical simulations of complex systems can be slow, they play an essential role for planning, operation, and control of, for example, fusion rectors and meta optics.
We are working on AI based short cuts -- developing surrogate models powered by knowledge about the underlying system with the potential for speeding up development cycles, making complex measurement data interpretable, and enabling real-time control in fusion reactor control rooms.
Collaborators:
Technical University of Denmark, Max-Planck Institute for Plasma Physics (Greifswald), and NILT
Contact person
 |
Jan-Matthias Braun, Associate professor SDU Applied AI and Data Science j-mb@mmmi.sdu.dk +45 65507892 |