Resolving Turbulent Magnetohydrodynamics: A Hybrid Operator-Diffusion Framework

TL;DR

This paper introduces a hybrid machine learning framework combining Physics-Informed Neural Operators and diffusion models to accurately simulate complex 2D MHD turbulence across a wide range of Reynolds numbers, capturing detailed spectral and statistical features.

Contribution

It presents a novel hybrid approach that integrates PINOs with diffusion models for high-fidelity turbulence simulation, extending surrogate modeling capabilities to extreme Reynolds numbers.

Findings

Achieves state-of-the-art accuracy in modeling MHD turbulence.

Successfully reconstructs spectral energy distributions and non-Gaussian statistics.

First surrogate to recover high-wavenumber magnetic field evolution at Re=10000.

Abstract

We present a hybrid machine learning framework that combines Physics-Informed Neural Operators (PINOs) with score-based generative diffusion models to simulate the full spatio-temporal evolution of two-dimensional, incompressible, resistive magnetohydrodynamic (MHD) turbulence across a broad range of Reynolds numbers ($\mathrm{Re}$). The framework leverages the equation-constrained generalization capabilities of PINOs to predict coherent, low-frequency dynamics, while a conditional diffusion model stochastically corrects high-frequency residuals, enabling accurate modeling of fully developed turbulence. Trained on a comprehensive ensemble of high-fidelity simulations with $\mathrm{Re} \in \{100, 250, 500, 750, 1000, 3000, 10000\}$, the approach achieves state-of-the-art accuracy in regimes previously inaccessible to deterministic surrogates. At $\mathrm{Re}=1000$ and $3000$, the model faithfully reconstructs the full spectral energy distributions of both velocity and magnetic fields late into the simulation, capturing non-Gaussian statistics, intermittent structures, and cross-field correlations with high fidelity. At extreme turbulence levels ($\mathrm{Re}=10000$), it remains the first surrogate capable of recovering the high-wavenumber evolution of the magnetic field, preserving large-scale morphology and enabling statistically meaningful predictions.