Learning Neural Operator Surrogates for the Black Hole Accretion Code

2026-04-28

AI Summary

This study addresses the high computational cost of general relativistic magnetohydrodynamic (GR-MHD) simulations of black hole accretion, which hinders systematic parameter studies. For the first time, physics-informed neural operators are applied to special relativistic resistive MHD (SRRMHD) and relativistic jet (SRMHD) scenarios to construct efficient surrogate models. Specifically, a physics-informed Fourier neural operator (PINO) accurately reproduces plasma blob structures in an unsupervised temporal setting, outperforming purely data-driven approaches. Additionally, OFormer—a Transformer-based architecture—effectively captures early-stage jet evolution features on high-resolution adaptive meshes for the first time. Both approaches enable rapid, physically consistent predictions that closely emulate results from BHAC simulations.

Abstract

General-relativistic magnetohydrodynamic (GR-MHD) simulations are essential for studying black hole accretion, relativistic jets, and magnetic reconnection, yet their computational cost severely limits systematic parameter exploration. We investigate neural operator surrogates for two astrophysically relevant simulation scenarios produced by the Black Hole Accretion Code (\texttt{BHAC}). First, a Physics Informed Fourier Neural Operator (PINO) is trained on the special-relativistic resistive MHD (SRRMHD) evolution of the Orszag-Tang vortex over a range of resistivities spanning the Sweet-Parker and fast reconnection regimes. By embedding the governing equations as an additional loss term evaluated at finer temporal resolution than the available data supervision, the model learns dynamics at time steps where no simulation data is provided, enabling recovery of plasmoid formation that a data-only baseline trained on the same sparse snapshots fails to reproduce. To our knowledge, the present work is the first application of a physics informed neural operator to special relativistic resistive MHD, and the first to investigate the capability of such models to resolve plasmoid formation in SRRMHD. In a second line of investigation, an OFormer-style Transformer Neural Operator is trained on the evolution of spine-sheath relativistic jets created with \texttt{BHAC}, in special-relativistic MHD (SRMHD). The model is directly applied on the adaptive mesh, highlighting the need for linear attention due to long sequences. The neural surrogate model is capable of capturing most of the major details, especially in early predictions. To our knowledge, this constitutes the first application of a neural operator directly on a high resolution adaptive mesh refinement grid in the context of MHD simulations.

Problem

• GR-MHD simulations
• computational cost
• parameter exploration
• black hole accretion
• relativistic jets

Innovation

• Physics-Informed Neural Operator
• Plasmoid Formation
• Adaptive Mesh Refinement
• Relativistic MHD
• Transformer Neural Operator