Abstract: Magnetic geometry has a significant effect on the level of turbulent transport in fusion plasmas. Here, we attempt to understand this dependence using multiple machine learning methods and a dataset of > 200,000 nonlinear simulations of ion-temperature-gradient turbulence in diverse non-axisymmetric geometries. At fixed gradients and other input parameters, the turbulent heat flux varies between geometries by many orders of magnitude. Patterns are apparent among the configurations with particularly high or particularly low heat flux. Regression and classification techniques from machine learning are then applied to extract patterns in the dataset. Due to a symmetry of the gyrokinetic equation, the heat flux and regressions thereof should be invariant to translations of the raw features in the parallel coordinate, similar to translation invariance in computer vision applications. Multiple regression models including convolutional neural networks (CNNs) and decision trees can achieve reasonable predictive power for the heat flux in held-out test configurations, with highest accuracy for the CNNs. Using sequential feature selection and Shapley values to measure feature importance, it is consistently found that the most important geometric lever on the heat flux is the flux surface compression in regions of bad curvature. The second most important geometric feature involves the magnitude of geodesic curvature. These two features align remarkably with surrogates that have been proposed recently based on theory, while the methods here allow a natural extension to more features for increased accuracy.

Talk time in other timezones:
AEDT  2:00 AM Fri 14 Feb,
JST  12:00 AM Fri 14 Feb,
CET   4:00 PM Thu 13 Feb,
GMT   3:00 PM Thu 13 Feb,
EST  10:00 AM Thu 13 Feb,
CST   9:00 AM Thu 13 Feb,
MST   8:00 AM Thu 13 Feb,
MST   7:00 AM Thu 13 Feb,
PST   7:00 AM Thu 13 Feb,
UTC  15:00    Thu 13 Feb