Gyrokinetic validation study using KSTAR plasmas

Abstract

To commercialize nuclear fusion energy, a sufficiently large number of fusion reactions must occur. High density and high temperature plasma should be confined long enough. To predict the performance of fusion plasma, understanding transport phenomena is essential. A validated transport model can predict the performance of future fusion devices including ITER and DEMO without expensive experiments. However, turbulence has been recognized as the main cause of transport in fusion plasma. Gyrokinetics is the most advanced tool to describe the dynamic of turbulence. Therefore, the validated gyrokinetic can predict the performance of future fusion plasmas, and as a result, it will be useful in designing future fusion reactors. Thus, the gyrokinetic validation study should be carried out for the commercialization of nuclear fusion. This thesis reports the progress in the first gyrokinetic validation study using KSTAR NBI heated L-mode plasmas. The experimental energy transport level was compared with the energy transport level calculated by gyrokinetic simulation. We focused on the quantitative comparison of energy flux between experiment and simulation at r/a=0.5. Uncertainty should be quantified for the quantitative comparison between experiment and simulation results. Uncertainties of input parameters for gyrokinetic analysis and experimental energy flux were quantified using profile samples and error propagation. CGYRO was used for gyrokinetic analysis. The linear stability analysis indicated that the most unstable mode was the trapped electron mode (TEM) at r/a=0.5 in the discharge studied in this thesis. The energy flux level calculated by gyrokinetic simulation was predicted to be lower than the experimental energy flux level estimated from power balance analysis. To check whether this result is valid or not, the energy flux was calculated by changing the values of the input parameters within their uncertainties, and it was estimated to be lower than the experimental energy flux. However, when the unidentified Z_{eff} value and the main thermal ion and impurity gradients were varied, and the energy flux predicted by the simulation was significantly changed. Therefore, the Z_{eff} value and the main thermal ion and impurity gradient are important parameters that affect the results of the gyrokinetic validation study, but they are unknown. As a result, it could not be concluded whether the gyrokinetic simulation well describes the experiment energy flux. However, this study confirmed that Z_{eff} profile information is required for the gyrokinetic validation study.