Experimental validation of a fast control-oriented, physics-informed surrogate model for plasma equilibrium reconstruction in the TCV tokamak
Abstract
Magnetic equilibrium reconstruction provides the plasma state estimate required for real-time shape control in tokamaks. We present a fast, physics-informed neural network surrogate of the \texttt{liuqe} equilibrium reconstruction code \cite{liuqe1} for the TCV tokamak at EPFL, achieving inference times below 100~μs and enabling 10~kHz shape control. The model is trained on around 10,000 TCV discharges spanning the full operational range of plasma shapes. Its modular branch/trunk architecture decouples magnetic measurement encoding from spatial coordinate processing, enabling physics-informed regularization via automatic differentiation of the predicted flux map. The surrogate has been compiled and deployed on the TCV real-time control system, and validated both offline and in real time against the models \texttt{liuqe-rt} and \texttt{lih}, showing comparable accuracy. Closed-loop performance assessed with the real-time software in-the-loop \texttt{fge} \cite{fge1} demonstrates control-equivalent behavior across multiple control strategies.
Related
- Real-time Tokamak Equilibrium Reconstruction Under Limited Experimental Data via Physics-Grounded Synthetic Pre-training
- Reconstruction of Unobserved Core Electron Density in Tokamak Plasmas with a Neural Operator Transport Surrogate and Uncertainty Quantification
- Parnassus for the CLD Detector: A Generative Machine-Learning Surrogate for Detector Simulation and Reconstruction at the FCC-ee
- CTrex: A Research-Oriented Framework for Kernel- and Projection-Level Algorithm Development in CT Reconstruction
- Diffusion prior for KSTAR equilibrium reconstruction under sensor dropout
BibTeX
@misc{grandin2026experimental,
title = {{Experimental validation of a fast control-oriented, physics-informed surrogate model for plasma equilibrium reconstruction in the TCV tokamak}},
author = {M. Grandin and A. Mele and C. Galperti and D. Gonzales Castineiras and C. Heiß and A. Cenedese and TCV team and EUROfusion Tokamak Exploitation team},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2606.09487},
eprint = {2606.09487},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.09487},
}
RIS
TY - UNPB
TI - Experimental validation of a fast control-oriented, physics-informed surrogate model for plasma equilibrium reconstruction in the TCV tokamak
AU - M. Grandin
AU - A. Mele
AU - C. Galperti
AU - D. Gonzales Castineiras
AU - C. Heiß
AU - A. Cenedese
AU - TCV team
AU - EUROfusion Tokamak Exploitation team
PY - 2026
DA - 2026/09/23
PB - arXiv
DO - 10.48550/arxiv.2606.09487
UR - https://arxiv.org/abs/2606.09487
AB - Magnetic equilibrium reconstruction provides the plasma state estimate required for real-time shape control in tokamaks. We present a fast, physics-informed neural network surrogate of the \texttt{liuqe} equilibrium reconstruction code \cite{liuqe1} for the TCV tokamak at EPFL, achieving inference times below 100~μs and enabling 10~kHz shape control. The model is trained on around 10,000 TCV discharges spanning the full operational range of plasma shapes. Its modular branch/trunk architecture decouples magnetic measurement encoding from spatial coordinate processing, enabling physics-informed regularization via automatic differentiation of the predicted flux map. The surrogate has been compiled and deployed on the TCV real-time control system, and validated both offline and in real time against the models \texttt{liuqe-rt} and \texttt{lih}, showing comparable accuracy. Closed-loop performance assessed with the real-time software in-the-loop \texttt{fge} \cite{fge1} demonstrates control-equivalent behavior across multiple control strategies.
ER -