Plasma surrogate modelling using Fourier neural operators

V Gopakumar, S Pamela, L Zanisi, Z Li, A Gray… - Nuclear …, 2024‏ - iopscience.iop.org
Predicting plasma evolution within a Tokamak reactor is crucial to realizing the goal of
sustainable fusion. Capabilities in forecasting the spatio-temporal evolution of plasma …

Data-driven models in fusion exhaust: AI methods and perspectives

S Wiesen, S Dasbach, A Kit, AE Jaervinen… - Nuclear …, 2024‏ - iopscience.iop.org
A review is given on the highlights of a scatter-shot approach of develo** machine-
learning methods and artificial neural networks based fast predictors for the application to …

Equivariant neural simulators for stochastic spatiotemporal dynamics

K Minartz, Y Poels, S Koop… - Advances in Neural …, 2023‏ - proceedings.neurips.cc
Neural networks are emerging as a tool for scalable data-driven simulation of high-
dimensional dynamical systems, especially in settings where numerical methods are …

Data efficiency and long term prediction capabilities for neural operator surrogate models of core and edge plasma codes

N Carey, L Zanisi, S Pamela, V Gopakumar… - arxiv preprint arxiv …, 2024‏ - arxiv.org
Simulation-based plasma scenario development, optimization and control are crucial
elements towards the successful deployment of next-generation experimental tokamaks and …

Multi-machine benchmark of the self-consistent 1D scrape-off layer model DIV1D from stagnation point to target with SOLPS-ITER

GL Derks, E Westerhof, M van Berkel… - Plasma Physics and …, 2024‏ - iopscience.iop.org
This paper extends a 1D dynamic physics-based model of the scrape-off layer (SOL)
plasma, DIV1D, to include the core SOL and possibly a second target. The extended model …

A novel Fourier neural operator framework for classification of multi-sized images: Application to three dimensional digital porous media

A Kashefi, T Mukerji - Physics of Fluids, 2024‏ - pubs.aip.org
Fourier neural operators (FNOs) are invariant with respect to the size of input images, and
thus images with any size can be fed into FNO-based frameworks without any modification …

NTVTOK-ML: Fast surrogate model for neoclassical toroidal viscosity torque calculation in tokamaks based on machine learning methods

XT Yan, NN Bao, CY Zhao, YW Sun, YT Meng… - Computer Physics …, 2025‏ - Elsevier
Abstract The Neoclassical Toroidal Viscosity (NTV) torque is a crucial source of toroidal
momentum in tokamaks, exerting significant influence on plasma instability and …

Prediction of fishbone linear instability in tokamaks with machine learning methods

ZY Liu, HR Qiu, GY Fu, Y **ao, YC Chen… - Nuclear …, 2024‏ - iopscience.iop.org
A machine learning based surrogate model for fishbone linear instability in tokamaks is
constructed. Hybrid simulations with the kinetic-magnetohydrodynamic (MHD) code M3D-K …

[HTML][HTML] Detachment scalings derived from 1D scrape-off-layer simulations

T Body, T Eich, A Kuang, T Looby, M Kryjak… - Nuclear Materials and …, 2024‏ - Elsevier
Fusion power plants will require detachment to mitigate sputtering and keep divertor heat
fluxes at tolerable levels. Controlling detachment on these devices may require the use of …

Accelerating Simulation of Two-Phase Flows with Neural PDE Surrogates

Y Poels, K Minartz, H Bansal, V Menkovski - arxiv preprint arxiv …, 2024‏ - arxiv.org
Simulation is a powerful tool to better understand physical systems, but generally requires
computationally expensive numerical methods. Downstream applications of such …