← Back to all news
arXiv Machine Learning By Daniel Gaytan-Villarreal, Aiken Xie, Tu Pham, Rohit Sonker, Chiara Amendola, Matteo Cremonesi, Cong Hao, Jeff Schneider

Real-Time Plasma State Prediction via FPGA-Accelerated Quantized Recurrent Probabilistic Neural Networks

Read the original on arXiv Machine Learning →

The paper presents an end‑to‑end workflow for deploying a recurrent probabilistic neural network (RPNN) on FPGA hardware to enable real‑time plasma state estimation for Tokamak control. By reducing architecture size and applying quantization‑aware training with QKeras, the model is synthesized with hls4ml for a Xilinx Alveo U50 device, achieving deterministic sub‑10 µs single‑timestep latency while staying within all four resource budgets (DSP, LUT, FF, BRAM).

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.

Related stories

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
More like this →
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-tuningbenchmarks
More like this →