TorbeamNN: machine learning-based steering of ECH mirrors on KSTAR
Authors:
- Andrew RothsteinARAndrew Rothstein
- Minseok KimMKMinseok Kim
- Minho WooMWMinho Woo
- Minsoo ChaMCMinsoo Cha
- Cheolsik ByunCBCheolsik Byun
- Sangkyeun KimSKSangkyeun Kim
- Keith EricksonKEKeith Erickson
- Youngho LeeYLYoungho Lee
- Josh Josephy-ZackJJJosh Josephy-Zack
- Jalal ButtJBJalal Butt
- Ricardo ShoushaRSRicardo Shousha
- Mi JoungMJMi Joung
Open Access
- 31 May 2025
- journal article
- Published by IOP Publishing in Plasma Physics and Controlled Fusion
- Vol. 67 (5) , 055036
Abstract
We have developed TorbeamNN: a machine learning surrogate model for the TORBEAM ray tracing code to predict electron cyclotron heating (ECH) 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 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 0.5 cm.Keywords
All Related Versions
Funding Information
- National Science Foundation Graduate Research Fellowship Program (DGE-2039656)
- Fusion Energy Sciences (DE- SC0015480)
This publication has 26 references indexed in Scilit:
- Design and operation results of KSTAR ECH systemFusion Engineering and Design, 2024
- Multi-chord IR–visible two-color interferometer on KSTARReview of Scientific Instruments, 2021
- Keras2c: A library for converting Keras neural networks to real-time compatible CEngineering Applications of Artificial Intelligence, 2021
- Accelerated version of NUBEAM capabilities in DIII-D using neural networksFusion Engineering and Design, 2021
- TORBEAM 2.0, a paraxial beam tracing code for electron-cyclotron beams in fusion plasmas for extended physics applicationsComputer Physics Communications, 2018
- Real-time beam tracing for control of the deposition location of electron cyclotron wavesFusion Engineering and Design, 2015
- Demonstration of sawtooth period control with EC waves in KSTAR plasmaEPJ Web of Conferences, 2015
- Overview of the ITER EC H&CD system and its capabilitiesFusion Engineering and Design, 2011
- Electron cyclotron emission diagnostics on KSTAR tokamakReview of Scientific Instruments, 2010
- Real time equilibrium reconstruction for tokamak discharge controlNuclear Fusion, 1998