M. Åqvist
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
M. Åqvist has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Aaron Ho
Massachusetts Institute of Technology
1 shared publication - Andreas Gillgren
Chalmers University of Technology, Gothenburg, Sweden
1 shared publication - Emil Fransson
Aix-Marseille Université, CNRS, PIIM, UMR 7345 , Marseille
1 shared publication J. Borsander
Chalmers University of Technology 1 , Gothenburg, Sweden
1 shared publication- O. Lindberg
Chalmers University of Technology 1 , Gothenburg, Sweden
1 shared publication - Par Strand
Chalmers University of Technology, Gothenburg, Sweden
1 shared publication
Latest Publications
- Open Access
- 1 December 2023
- Published by AIP Publishing in Physics of Plasmas
- Vol. 30 (12)
A fast neural network surrogate model for the eigenvalues of QuaLiKiz
We introduce a neural network surrogate model that predicts the eigenvalues for the turbulent microinstabilities, based on the gyrokinetic eigenvalue solver in QuaLiKiz. The model quickly provides information about the dominant instability for specific plasma conditions, and in addition, the eigenvalues offer a pathway for extrapolating transport fluxes. The model is trained on a 5 × 106 data points large dataset based on experimental data from discharges at the joint European torus, where each data point represents a QuaLiKiz simulation. The most accurate model was obtained when the task was split into a classification task to decide if the imaginary part of eigenvalues were stable (≤0) or not, and a regression model to calculate the eigenvalues once the classifier predicted the unstable class.
- 1 December 2023
- Published by AIP Publishing in Physics of Plasmas
- Vol. 30 (12)
Open AccessA fast neural network surrogate model for the eigenvalues of QuaLiKiz
We introduce a neural network surrogate model that predicts the eigenvalues for the turbulent microinstabilities, based on the gyrokinetic eigenvalue solver in QuaLiKiz. The model quickly provides information about the dominant instability for specific plasma conditions, and in addition, the eigenvalues offer a pathway for extrapolating transport fluxes. The model is trained on a 5 × 106 data points large dataset based on experimental data from discharges at the joint European torus, where each data point represents a QuaLiKiz simulation. The most accurate model was obtained when the task was split into a classification task to decide if the imaginary part of eigenvalues were stable (≤0) or not, and a regression model to calculate the eigenvalues once the classifier predicted the unstable class.