arXiv Machine Learning July 2, 2026 By Alasdair Ross, George K. Holt, Kamran Pentland, Adriano Agnello, Nicola C. Amorisco, Pedro Cavestany, Aran Garrod, Timothy Nunn, Charles Vincent, Graham McArdle

Real-time virtual circuits for plasma shape control via neural network emulators

Read the original on arXiv Machine Learning →

arXiv:2605. 14939v2 Announce Type: replace-cross Abstract: Reliable position and shape control in tokamak plasmas requires accurate real-time regulation of several strongly coupled shape parameters.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Aug 28

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

The paper reports on the integration and testing of neural‑network emulators that predict plasma shape for the MAST‑Upgrade tokamak’s plasma control system. The neural models use plasma current, poloidal field coil currents, and plasma profile parameters to output shape predictions and Jacobians, which are then used by a real‑time C++ inference server to compute virtual circuit matrices and updated coil current requests. The work emphasizes a validation workflow and best practices to build confidence in the AI‑based control framework before experimental deployment.

By Matthew J. Marshall, Edward Jones, Graham J. McArdle, Alasdair Ross, Kamran Pentland, Nicola C. Amorisco, Charles Vincent, Martin Kochan, Colin Hogben, Graham Jones, Adam Stephen, George K. Holt, Adriano Agnello

arXiv AI

Aug 25

AI Surrogate Modeling for Real-Time Tokamak Equilibrium Prediction: Benchmarking Neural Architectures and Validation on EXL-50U

arXiv:2608.23217v1 Announce Type: cross Abstract: Fast and reliable plasma equilibrium prediction is essential for real-time tokamak operation and control, but conventional Grad-Shafranov (GS) solver...

By Guoyang Shi, Zitong Zhang, Siqi Ding, Jianguo Chen, Yapeng Zhang, Jiayi Zhi, Hanyue Zhao, Tianyuan Liu

llms benchmarks

arXiv Machine Learning

Jun 16

Towards Data-Efficient Cross-Device Generalization of Grad-Shafranov Equilibria via Transfer Learning Neural Operator

arXiv:2606. 15512v1 Announce Type: new Abstract: Real-time reconstruction of magnetohydrodynamic equilibria is essential for plasma shaping, stability assessment and feedback control in magnetic confinement fusion.

By Jay Phil Yoo, William Howes, Yashika Ghai, Kazuma Kobayashi, Souvik Chakraborty, Syed Bahauddin Alam

fine-tuning benchmarks

Hugging Face Trending Papers

Aug 24

AI Surrogate Modeling for Real-Time Tokamak Equilibrium Prediction: Benchmarking Neural Architectures and Validation on EXL-50U

Fast and reliable plasma equilibrium prediction is essential for real-time tokamak operation and control, but conventional Grad-Shafranov (GS) solvers are often too costly for real-time deployment. We...

llms benchmarks

arXiv Machine Learning

Jun 10

Offline Reinforcement Learning for Rotation Profile Control in Tokamaks

arXiv:2605. 05857v2 Announce Type: replace Abstract: Tokamaks remain leading candidates for achieving practical fusion energy, yet many important control problems inside these devices are still difficult or unsolved.

By Rohit Sonker, Hiro Josep Farre Kaga, Jiayu Chen, Andrew Rothstein, Ian Char, Ricardo Shousha, Egemen Kolemen, Jeff Schneider

reinforcement-learning

arXiv AI

Jul 24

Cycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas

arXiv:2607. 21407v1 Announce Type: cross Abstract: The boundary and divertor plasma govern how a tokamak exhausts power and particles, setting heat fluxes, target conditions, and the onset of detachment.

By Abdourahmane Diaw, Sebastian De Pascuale, Jae-Sun Park, Ivan Paradela Perez, Jeremy D. Lore, Stefan Dasbach

One email a morning, machine-written

Related stories

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

AI Surrogate Modeling for Real-Time Tokamak Equilibrium Prediction: Benchmarking Neural Architectures and Validation on EXL-50U

Towards Data-Efficient Cross-Device Generalization of Grad-Shafranov Equilibria via Transfer Learning Neural Operator

AI Surrogate Modeling for Real-Time Tokamak Equilibrium Prediction: Benchmarking Neural Architectures and Validation on EXL-50U

Offline Reinforcement Learning for Rotation Profile Control in Tokamaks

Cycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas