Magnetohydrodynamics with Physics Informed Neural Operators
arXiv:2302.08332 · doi:10.1088/2632-2153/ace30a
Abstract
The modeling of multi-scale and multi-physics complex systems typically involves the use of scientific software that can optimally leverage extreme scale computing. Despite major developments in recent years, these simulations continue to be computationally intensive and time consuming. Here we explore the use of AI to accelerate the modeling of complex systems at a fraction of the computational cost of classical methods, and present the first application of physics informed neural operators to model 2D incompressible magnetohydrodynamics simulations. Our AI models incorporate tensor Fourier neural operators as their backbone, which we implemented with the TensorLY package. Our results indicate that physics informed neural operators can accurately capture the physics of magnetohydrodynamics simulations that describe laminar flows with Reynolds numbers . We also explore the applicability of our AI surrogates for turbulent flows, and discuss a variety of methodologies that may be incorporated in future work to create AI models that provide a computationally efficient and high fidelity description of magnetohydrodynamics simulations for a broad range of Reynolds numbers. The scientific software developed in this project is released with this manuscript.
arXiv comment32 pages, 24 figures, 1 table. First application of physics informed neural operators to solve magnetohydrodynamics equations, v2: Accepted to Machine Learning: Science and Technology
References in corpus (28)
- Decoupled Weight Decay Regularization
- DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators
- Gaussian Error Linear Units (GELUs)
- Fourier Neural Operator for Parametric Partial Differential Equations
- Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations
- Dedalus: A Flexible Framework for Numerical Simulations with Spectral Methods
- Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations
- Efficient magnetic-field amplification due to the Kelvin-Helmholtz instability in binary neutron star mergers
- Multipole Graph Neural Operator for Parametric Partial Differential Equations
- Physics-Informed Neural Operator for Learning Partial Differential Equations
- Universal Nonlinear Small-Scale Dynamo
- Turbulent magnetic field amplification in binary neutron star mergers
- Energy transfer in compressible magnetohydrodynamic turbulence
- MHD Turbulence
- Attention-Enhanced Neural Network Models for Turbulence Simulation
- Binary Neutron Star Merger Simulations with a Calibrated Turbulence Model
- Applications of physics informed neural operators
- Universality of the turbulent magnetic field in hypermassive neutron stars produced by binary mergers
- A Constrained Transport Scheme for MHD on Unstructured Static and Moving Meshes
- GRMHD large eddy simulations with gradient subgrid-scale model
- Gradient sub-grid-scale model for relativistic MHD Large Eddy Simulations
- A nonlinear structural subgrid-scale closure for compressible MHD Part II: a priori comparison on turbulence simulation data
- Extension of the sub-grid-scale gradient model for compressible magnetohydrodynamics turbulent instabilities
- Comparative statistics of selected subgrid-scale models in large eddy simulations of decaying, supersonic MHD turbulence
- Artificial neural network subgrid models of 2-D compressible magnetohydrodynamic turbulence
- A nonlinear structural subgrid-scale closure for compressible MHD Part I: derivation and energy dissipation properties
- Nonlinear closures for scale separation in supersonic magnetohydrodynamic turbulence
- Physics-Informed Machine Learning for Modeling Turbulence in Supernovae
Cited by in corpus (5)
- Long-term predictions of turbulence by implicit U-Net enhanced Fourier neural operator
- Applications of machine learning in gravitational wave research with current interferometric detectors
- Resolving Turbulent Magnetohydrodynamics: A Hybrid Operator-Diffusion Framework
- Modeling turbulent and self-gravitating fluids with Fourier neural operators
- APRIL: Auxiliary Physically-Redundant Information in Loss -- A physics-informed framework for parameter estimation with a gravitational-wave case study