Department of Physics, Aalto University School of Science
Local Project ID:
HPC Platform used:
SuperMUC of LRZ
In order to invest in a sustainable energy mix and avoid CO2 emissions, new energy sources such as fusion energy need to be developed. Understanding turbulent transport is needed for further optimization of fusion reactors but realistic transport time scale simulations of plasma turbulence are computationally very demanding. The aim of the present project is to increase the understanding of the mechanisms behind the sudden improvement in confinement observed in experiments.
The research activities carried out on GCS supercomputer SuperMUC of LRZ Garching shed some light on the possible role of the radial derivative of a time-varying electric field in triggering this transition using the ELMFIRE turbulence simulation code which investigates these phenomena with a so-called first principal computer model. This model tracks individual particles providing information on the complex interplay between the magnetic field, the electric field and the particle trajectories.
ELMFIRE is even more computationally demanding than other turbulence code as it does not assume a Maxwellian distribution of particles but instead simulates the full distribution of electrons and ions. While this approach enables more realistic simulations especially near the plasma edge, a very large number of simulation particles are needed for good statistics.
So far excellent results for the small FT-2 tokamak have been obtained and the GCS supercomputing resources, made available through the Partnership for Advanced Computing in Europe (PRACE), were used to investigate phenomenology of confinement improvement in the mid-sized tokamak TEXTOR.
The main effort in the project was a scan over local parameters like temperature and density starting from experimental Textor parameters. From this data, scientists have managed to interpret old Textor experiments and also cross-validate a recent sophisticated analytic theory of radial wave length of radial electric field oscillation. With the present data researchers are also able to investigate the parametric dependence of the amplitude of the oscillation. From these three quantities one can derive an equation for the contribution of time varying electric field on turbulence suppression which can be one important part in understanding the sudden improvement in confinement. As confinement improves when building larger fusion devices, theoretical understanding of optimizing confinement helps to build cheaper reactors in the future.
The scientific team included Timo Kiviniemi (project leader), Paavo Niskala, Susan Leerink, Salomon Janhunen (now at Brown University) and Tuomas Korpilo from Aalto University and Jukka Heikkinen from VTT. Experimental contact in Textor was Andreas Krämer-Flecken. Optimization support was received from Artur Signell and Jan Westerholm from the Åbo Akademi.
This project was made possible by PRACE (Partnership for Advanced Computing in Europe) granting computing time on HPC system SuperMUC of LRZ.
Department of Applied Physics
Aalto University School of Science
P.O.Box 14100, FI-00076 Aalto/Finland
LRZ project ID: pr86go