Brief

Researchers have developed a novel neural operator framework to accelerate the real-time reconstruction of magnetohydrodynamic equilibria in fusion devices. This approach recasts equilibrium reconstruction as a cross-device operator learning problem, mapping geometry and profile parameters directly to the poloidal flux field. The Wavelet Neural Operator architecture demonstrated strong cross-geometry performance, achieving low relative L2 errors with limited labeled data and enabling millisecond-scale inference. AI

Researchers have developed neural network emulators to generate virtual circuits for real-time plasma shape control in tokamak devices. This approach uses a large dataset of simulated equilibria to train emulators that can rapidly derive accurate control vectors. The method aims to improve upon traditional techniques that rely on precomputed virtual circuits, which can degrade in performance as plasma configurations evolve. AI