Real-time virtual circuits for plasma shape control via neural network emulators: integration and testing in the MAST-U PCS

arXiv:2608.26216 · physics.plasm-ph, cs.LG · Submitted 2026-08-26 · Read on arXiv

physics.plasm-ph, cs.LG

Submitted: 2026-08-26

Updated: 2026-08-26

Comments: 8 pages, 3 figures, presented at 15th Technical Meeting on Control Systems, Data Acquisition, Data Management and Remote Participation in Fusion Research and submitted for review at Fusion Engineering and Design on July 31 2026

Code: https://github.com/Tencent/ncnn

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Accuracy — 5× in this paper · explained in 1 episode
• Core — 4× in this paper · explained in 2 episodes
• Reinforcement Learning — 3× in this paper · explained in 4 episodes
• Calibration — 2× in this paper · explained in 7 episodes
• Deep Reinforcement Learning — 2× in this paper · explained in 2 episodes
• Adam — 1× in this paper · explained in 2 episodes
• Adaptivity — 1× in this paper · explained in 1 episode
• Communication Overhead — 1× in this paper · explained in 2 episodes
• FreeGSNKE Pulse Design Tool — 1× in this paper
• Robustness — 1× in this paper · explained in 5 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• Real-time virtual circuits for plasma shape control via neural network emulators: dynamic validation in closed-loop simulations
• The FreeGSNKE Pulse Design Tool (FPDT): a computational framework for evolutive plasma scenario and control design

Related papers

• Reconstruction of Multiscale Plasma Dynamics Across Operating Regimes
• Is Your AI Fast Enough to Run a Fusion Reactor?
• GyroSwin: 5D Surrogates for Gyrokinetic Plasma Turbulence Simulations
• Synthetic 2D magnetic turbulence with controlled island structuring
• Electromagnetic ghosts in pair plasmas
• Collisionless whistler heat-flux instability in ultra-high- beta plasmas