TorbeamNN: Machine learning based steering of ECH mirrors on KSTAR
Abstract
We have developed TorbeamNN: a machine learning surrogate model for the TORBEAM ray tracing code to predict electron cyclotron heating and current drive locations in tokamak plasmas. TorbeamNN provides more than a 100 times speed-up compared to the highly optimized and simplified real-time implementation of TORBEAM without any reduction in accuracy compared to the offline, full fidelity TORBEAM code. The model was trained using KSTAR electron cyclotron heating (ECH) mirror geometries and works for both O-mode and X-mode absorption. The TorbeamNN predictions have been validated both offline and real-time in experiment. TorbeamNN has been utilized to track an ECH absorption vertical position target in dynamic KSTAR plasmas as well as under varying toroidal mirror angles and with a minimal average tracking error of 0.5cm.
Keywords
Cite
@article{arxiv.2504.11648,
title = {TorbeamNN: Machine learning based steering of ECH mirrors on KSTAR},
author = {Andrew Rothstein and Minseok Kim and Minho Woo and Minsoo Cha and Cheolsik Byun and Sangkyeun Kim and Keith Erickson and Youngho Lee and Josh Josephy-Zack and Jalal Butt and Ricardo Shousha and Mi Joung and June-Woo Juhn and Kyu-Dong Lee and Egemen Kolemen},
journal= {arXiv preprint arXiv:2504.11648},
year = {2026}
}
Comments
7 pages, 5 figures