Machine learning of hidden variables in multiscale fluid simulation
AS Joglekar, AGR Thomas - Machine Learning: Science and …, 2023 - iopscience.iop.org
Solving fluid dynamics equations often requires the use of closure relations that account for
missing microphysics. For example, when solving equations related to fluid dynamics for …
missing microphysics. For example, when solving equations related to fluid dynamics for …
EMWP-RNN: A Physics-Encoded Recurrent Neural Network for Wave Propagation in Plasmas
Electromagnetic (EM) wave propagation and inversion in complex time-varying medium is a
challenging problem, particularly for plasma applications. We extend the EM wave–plasma …
challenging problem, particularly for plasma applications. We extend the EM wave–plasma …
Data-driven modeling of Landau damping by physics-informed neural networks
Kinetic approaches are generally accurate in dealing with microscale plasma physics
problems but are computationally expensive for large-scale or multiscale systems. One of …
problems but are computationally expensive for large-scale or multiscale systems. One of …
A novel ionospheric inversion model: PINN‐SAMI3 (physics informed neural network based on SAMI3)
Purely data‐driven ionospheric modeling fails to adequately obey fundamental physical
laws. To overcome this shortcoming, we propose a novel ionospheric inversion model …
laws. To overcome this shortcoming, we propose a novel ionospheric inversion model …
[HTML][HTML] Numerical study of magnetic island coalescence using magnetohydrodynamics with adaptively embedded particle-in-cell model
Collisionless magnetic reconnection typically requires kinetic treatment that is, in general,
computationally expensive compared to fluid-based models. In this study, we use the …
computationally expensive compared to fluid-based models. In this study, we use the …
Fast forward modeling of magnetotelluric data in complex continuous media using an extended Fourier DeepONet architecture
The calculations of magnetotelluric (MT) responses play a fundamental role in the inversion
and resolution analysis for MT problems. Conventional numerical methods for forward …
and resolution analysis for MT problems. Conventional numerical methods for forward …
Electron cyclotron drift instability and anomalous transport: two-fluid moment theory and modeling
In the presence of a strong electric field perpendicular to the magnetic field, the electron
cross-field (E× B) flow relative to the unmagnetized ions can cause the so-called electron …
cross-field (E× B) flow relative to the unmagnetized ions can cause the so-called electron …
Data-driven discovery of a heat flux closure for electrostatic plasma phenomena
ER Ingelsten, MC McGrae-Menge, EP Alves… - arXiv preprint arXiv …, 2024 - arxiv.org
Progress in understanding multi-scale collisionless plasma phenomena requires employing
tools which balance computational efficiency and physics fidelity. Collisionless fluid models …
tools which balance computational efficiency and physics fidelity. Collisionless fluid models …
Data-driven Modeling of Plasma Fluid Closure and Parameter Prediction
K Liu, W Cheng, H Fu, M Jiang, J Ma… - 2022 International …, 2022 - ieeexplore.ieee.org
The advances of technology enable us to discover partial differential equations (PDEs) from
data of a certain size, which is of great significance for modeling physical systems. In this …
data of a certain size, which is of great significance for modeling physical systems. In this …
Data-Driven Modeling of Landau Damping by Fourier Neural Operator
The development of machine learning techniques enables us to discover partial differential
equations from sparse data, which has important implications for modeling complex physical …
equations from sparse data, which has important implications for modeling complex physical …