High temporal resolution of pedestal dynamics via machine learning on density diagnostics
Journal article, 2024

At the Joint European Torus, the reference diagnostic to measure electron density is Thomson scattering. However, this diagnostic has a low sampling rate, which makes it impractical to study the temporal dynamics of fast processes, such as edge localized modes. In this work, we use machine learning to predict the density profile based on data from another diagnostic, namely reflectometry. By learning to transform reflectometry data into Thomson scattering profiles, the model is able to generate the density profile at a much higher sampling rate than Thomson scattering, and more accurately than reflectometry alone. This enables the study of pedestal dynamics, by analyzing the time evolution of the pedestal height, width, position and gradient. We also discuss the accuracy of the model when applied on experimental campaigns that are different from the one it was trained on.

machine learning

reflectometry

plasma diagnostics

thomson scattering

Author

Diogo R. Ferreira

University of Lisbon

Andreas Gillgren

Chalmers, Space, Earth and Environment, Astronomy and Plasmaphysics

Andrei Osipov

Chalmers, Space, Earth and Environment, Astronomy and Plasmaphysics

Pär Strand

Chalmers, Space, Earth and Environment, Astronomy and Plasmaphysics

JET contributors

Plasma Physics and Controlled Fusion

0741-3335 (ISSN) 1361-6587 (eISSN)

Vol. 66 2 025001

Implementation of activities described in the Roadmap to Fusion during Horizon Europe through a joint programme of the members of the EUROfusion consortium

European Commission (EC) (101052200), 2021-01-01 -- 2025-12-31.

Infrastruktur för forskning och utveckling av fusionsreaktorer: ITER och DEMO

Swedish Research Council (VR) (2021-00182), 2022-01-01 -- 2026-12-31.

Subject Categories

Fusion, Plasma and Space Physics

DOI

10.1088/1361-6587/ad15ef

More information

Latest update

1/9/2024 1