Autoregressive prediction of 2D MHD dynamics inferred from deep learning modeling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kivarkis, David, Mouhali, Waleed, Benkadda, Sadruddin, Schneider, Kai
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914491810185216
author Kivarkis, David
Mouhali, Waleed
Benkadda, Sadruddin
Schneider, Kai
author_facet Kivarkis, David
Mouhali, Waleed
Benkadda, Sadruddin
Schneider, Kai
contents We develop two deep learning surrogate autoregressive models for the prediction of the temporal evolution of two-dimensional ideal magnetohydrodynamic (MHD) Kelvin-Helmholtz instabilities across a range of magnetic field strengths. Using two neural network architectures, a Koopman-based Transformer model and a ConvLSTM-UNet, our approach enables simultaneous prediction of vorticity and current density directly from high-resolution simulations. The models are trained in an autoregressive manner and are able to reproduce key features of the multiscale dynamics over several instability growth and nonlinear saturation phases. Beyond accurate field reconstruction, the surrogates preserve essential physical structures of ideal MHD dynamics, including the conservation trends of global invariants and the propagation of Alfvénic fluctuations. Compared to direct numerical simulations, the proposed surrogates offer substantially reduced computational cost while maintaining good agreement with the reference dynamics. These results suggest that deep learning based surrogate models can provide a promising complementary tool for the efficient and physically consistent exploration of high-fidelity plasma and fluid simulations.
format Preprint
id arxiv_https___arxiv_org_abs_2604_18221
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Autoregressive prediction of 2D MHD dynamics inferred from deep learning modeling
Kivarkis, David
Mouhali, Waleed
Benkadda, Sadruddin
Schneider, Kai
Plasma Physics
Computational Physics
Fluid Dynamics
76W05, 76Exx, 68T07
We develop two deep learning surrogate autoregressive models for the prediction of the temporal evolution of two-dimensional ideal magnetohydrodynamic (MHD) Kelvin-Helmholtz instabilities across a range of magnetic field strengths. Using two neural network architectures, a Koopman-based Transformer model and a ConvLSTM-UNet, our approach enables simultaneous prediction of vorticity and current density directly from high-resolution simulations. The models are trained in an autoregressive manner and are able to reproduce key features of the multiscale dynamics over several instability growth and nonlinear saturation phases. Beyond accurate field reconstruction, the surrogates preserve essential physical structures of ideal MHD dynamics, including the conservation trends of global invariants and the propagation of Alfvénic fluctuations. Compared to direct numerical simulations, the proposed surrogates offer substantially reduced computational cost while maintaining good agreement with the reference dynamics. These results suggest that deep learning based surrogate models can provide a promising complementary tool for the efficient and physically consistent exploration of high-fidelity plasma and fluid simulations.
title Autoregressive prediction of 2D MHD dynamics inferred from deep learning modeling
topic Plasma Physics
Computational Physics
Fluid Dynamics
76W05, 76Exx, 68T07
url https://arxiv.org/abs/2604.18221