Abstract: In this talk I will first briefly introduce the concept of fusion energy, and discuss the pathways towards enabling fusion as a viable source of electricity production. After overviewing the progress to date, I will present the key remaining plasma and materials physcs challenges towards enabling reliable energy production in tokamaks, the furthest developed fusion concept.

I will then discuss why computational approaches are crucial for design of fusion power plants, and discuss where and how machine learning approaches can give the otherwise extremely heavy simulations a boost, either in terms of computational efficiency or model accuracy.

Speakers:  Kai Nordlund is professor of computational materials physics and dean of the Faculty of Science at the University of Helsinki. He received his PhD in physics in 1995 at the University of Helsinki, and after postdoc positions at the University of Illinois and Academy of Finland was appointed full professor at his alma mater in 2003. Now he is leading as dean a Faculty with more than 1000 employees, and as professor a 15-person research group doing quantum mechanical, classical and mesoscale atomistic simulations of radiation and other non-equilibrium effects in all classes of materials. As of 2022, he has published more than 560 refereed publications, and his h-index exceeds his age

Affiliation: University of Helsinki

Place of Seminar:  Zoom