CARONTE: A Physics-Informed Extreme Learning Machine-Based Algorithm for Plasma Boundary Reconstruction in Magnetically Confined Fusion Devices
IEEE Access · 2026 · 14 · 105705-105721
DOI: 10.1109/access.2026.3712498 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Federico Fiorenza und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
In this work, we propose a novel physics-informed neural network-based algorithm for real-time plasma boundary reconstruction in tokamak devices. The approach is based on a single Extreme Learning Machine network used to solve the homogeneous Grad–Shafranov equation, which is required to identify the plasma boundary. This architecture enables real-time training of the network parameters using the available magnetic sensor data, allowing the network output to adapt dynamically to the evolving plasma equilibrium. We demonstrate that the network performs accurate plasma boundary reconstruction for complex configurations, outperforming well-established methods, such as the algorithm used for decades at the Joint European Torus, the world’s largest tokamak, until it ceased operation in 2023. Indeed, compared to the latter, the proposed solution better generalizes the poloidal flux function, without requiring algorithm retuning across different plasma equilibria. The proposed neural reconstructor also demonstrates greater robustness with respect to noise on the magnetic measurements. Furthermore, the proposed approach exploits Bayesian optimization to automate the tuning of the network’s non-interpretable hyperparameters. This method takes advantage of the generalization power of neural networks but without the need for extensive, time-consuming training based on a huge amount of experimental data, making its implementation on existing devices straightforward. Finally, we prove that the proposed algorithm is suitable for implementation on existing fusion devices, by assessing its performance on a real-time target.
Zitieren
APA
Fiorenza, F., Dubbioso, S., Tommasi, G. D., & Pironti, A. (2026). CARONTE: A Physics-Informed Extreme Learning Machine-Based Algorithm for Plasma Boundary Reconstruction in Magnetically Confined Fusion Devices. IEEE Access, 14, 105705-105721. https://doi.org/10.1109/access.2026.3712498
Vancouver
Fiorenza F, Dubbioso S, Tommasi GD, Pironti A. CARONTE: A Physics-Informed Extreme Learning Machine-Based Algorithm for Plasma Boundary Reconstruction in Magnetically Confined Fusion Devices. IEEE Access. 2026;14:105705-105721. doi:10.1109/access.2026.3712498
BibTeX
@article{Fiorenza2026,
author = {Fiorenza, Federico and Dubbioso, Sara and Tommasi, Gianmaria De and Pironti, Alfredo},
title = {CARONTE: A Physics-Informed Extreme Learning Machine-Based Algorithm for Plasma Boundary Reconstruction in Magnetically Confined Fusion Devices},
journal = {IEEE Access},
year = {2026},
volume = {14},
pages = {105705--105721},
doi = {10.1109/access.2026.3712498}
}
EUVIMED