Real-time estimation and control of the electron density with a novel multi-rate observer on TCV
POSTER
Abstract
The estimation and control in real time of the electron density for tokamaks is fundamental for its operation in many experimental conditions and scenarios. A precise assessment of the electron density profile during the discharge enables a more accurate control of various derived quantities, such as core, edge and line-averaged electron density within the LCFS.
Building on the integration of RAPDENS into the TCV plasma control system [1], this follow-up study explores its application to density control for detachment studies and high-performance H-mode scenarios.
The RAPDENS-based density profile observer combines the spatial resolution of Thomson Scattering with the high time resolution of the FIR to obtain a reliable density profile estimate, which is then employed to control the normalized edge density [2] for advanced tokamak scenarios in H-mode plasmas.
Experiments demonstrated the observer’s capability to support detachment studies in complex divertor geometries by accurately controlling the line-averaged density within the LCFS while avoiding divertor pick-up of the FIR signal.
Unknown plasma parameters can be estimated and adapted in real-time through the Kalman filter algorithm, such as fuelling efficiency and particle transport coefficients, for enhanced model robustness.
[1] F. Pastore et al., Fus. Eng. Des vol.192, p.113615,doi:10.1016/j.fusengdes.2023.113615, 2023.
[2] M. Bernert et al 2015 Plasma Phys. Control. Fusion 57 014038, doi: 10.1088/0741-3335/57/1/014038
Building on the integration of RAPDENS into the TCV plasma control system [1], this follow-up study explores its application to density control for detachment studies and high-performance H-mode scenarios.
The RAPDENS-based density profile observer combines the spatial resolution of Thomson Scattering with the high time resolution of the FIR to obtain a reliable density profile estimate, which is then employed to control the normalized edge density [2] for advanced tokamak scenarios in H-mode plasmas.
Experiments demonstrated the observer’s capability to support detachment studies in complex divertor geometries by accurately controlling the line-averaged density within the LCFS while avoiding divertor pick-up of the FIR signal.
Unknown plasma parameters can be estimated and adapted in real-time through the Kalman filter algorithm, such as fuelling efficiency and particle transport coefficients, for enhanced model robustness.
[1] F. Pastore et al., Fus. Eng. Des vol.192, p.113615,doi:10.1016/j.fusengdes.2023.113615, 2023.
[2] M. Bernert et al 2015 Plasma Phys. Control. Fusion 57 014038, doi: 10.1088/0741-3335/57/1/014038
Presenters
-
Francesco Pastore
EPFL, École Polytechnique Fédérale de Lausanne
Authors
-
Francesco Pastore
EPFL, École Polytechnique Fédérale de Lausanne
-
Olivier Sauter
EPFL, SPC-EPFL, Ecole Polytechnique Federale de Lausanne
-
Federico Felici
Google DeepMind
-
Daniela Kropackova
Czech Technical University, Prague
-
Ondrej Kudlacek
Max-Planck-Institut für Plasmaphysik
-
N. M. T. Vu
ITER Organization
-
Alessandro Pau
École Polytechnique Fédérale de Lausanne, SPC-EPFL, École Polytechnique Fédérale de Lausanne (EPFL), Swiss Plasma Center (SPC)
-
Cristian Galperti
SPC-EPFL, Ecole Polytechnique Federale de Lausanne
-
Timo Ravensbergen
ITER Organization
-
Simon Van Mulders
ITER Organization
-
Kenneth Lee
SPC-EPFL
-
Benjamin Vincent
SPC-EPFL