Data-driven modeling of Landau damping by physics-informed neural networks

Abstract

Kinetic approaches are generally accurate in dealing with microscale plasma physics problems but are computationally expensive for large-scale or multiscale systems. One of the long-standing problems in plasma physics is the integration of kinetic physics into fluid models, which is often achieved through sophisticated analytical closure terms. In this paper, we successfully construct a multi-moment fluid model with an implicit fluid closure included in the neural network using machine learning. The multi-moment fluid model is trained with a small fraction of sparsely sampled data from kinetic simulations of Landau damping, using the physics-informed neural network (PINN) and the gradient-enhanced physics-informed neural network (gPINN). The multi-moment fluid model constructed using either PINN or gPINN reproduces the time evolution of the electric field energy, including its damping rate, and the plasma dynamics from the kinetic simulations. In addition, we introduce a variant of the gPINN architecture, namely, gPINN$p$ to capture the Landau damping process. Instead of including the gradients of all the equation residuals, gPINN$p$ only adds the gradient of the pressure equation residual as one additional constraint. Among the three approaches, the gPINN$p$-constructed multi-moment fluid model offers the most accurate results. This work sheds light on the accurate and efficient modeling of large-scale systems, which can be extended to complex multiscale laboratory, space, and astrophysical plasma physics problems.

arXiv comment11 pages, 7 figures, accepted for publication in Physical Review Research

References in corpus (20)

• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Gradient-enhanced physics-informed neural networks for forward and inverse PDE problems
• PDE-Net 2.0: Learning PDEs from Data with A Numeric-Symbolic Hybrid Deep Network
• Flow over an espresso cup: Inferring 3D velocity and pressure fields from tomographic background oriented schlieren videos via physics-informed neural networks
• Data-driven rogue waves and parameter discovery in the defocusing NLS equation with a potential using the PINN deep learning
• Discontinuous Galerkin algorithms for fully kinetic plasmas
• Uncovering turbulent plasma dynamics via deep learning from partial observations
• Global Ten-Moment Multifluid Simulations of the Solar Wind Interaction with Mercury: From the Planetary Conducting Core to the Dynamic Magnetosphere
• Data-driven discovery of reduced plasma physics models from fully-kinetic simulations
• Electron Physics in 3D Two-Fluid Ten-Moment Modeling of Ganymede's Magnetosphere
• Relativistic Magnetic Reconnection in the Laboratory
• Machine Learning Surrogate Models for Landau Fluid Closure
• Exact and Locally Implicit Source Term Solvers for Multifluid-Maxwell Systems
• An introductory guide to fluid models with anisotropic temperatures Part 2 -- Kinetic theory, Padé approximants and Landau fluid closures
• A Six-moment Multi-fluid Plasma Model
• Data-driven, multi-moment fluid modeling of Landau damping
• Identification of high order closure terms from fully kinetic simulations using machine learning
• Numerical Study of Magnetic Island Coalescence Using Magnetohydrodynamics With Adaptively Embedded Particle-In-Cell Model
• High Flux Femtosecond X-ray Emission from the Electron-Hose Instability in Laser Wakefield Accelerators
• Discovery of double Hall pattern associated with collisionless magnetic reconnection in dusty plasmas

Cited by in corpus (8)

• Machine-learning heat flux closure for multi-moment fluid modeling of nonlinear Landau damping
• Numerical Study of Magnetic Island Coalescence Using Magnetohydrodynamics With Adaptively Embedded Particle-In-Cell Model
• Data-driven discovery of a heat flux closure for electrostatic plasma phenomena
• Machine-learning Closure for Vlasov-Poisson Dynamics in Fourier-Hermite Space
• Surrogate Modeling of Landau Damping with Deep Operator Networks
• Embedding physical symmetries into machine-learned reduced plasma physics models via data augmentation
• The Machine Learning Approach to Moment Closure Relations for Plasma: A Review
• Neural-network-based longitudinal electric field prediction in nonlinear plasma wakefield accelerators