Predicting continuum breakdown with deep neural networks
From MaRDI portal
Abstract: The multi-scale nature of gaseous flows poses tremendous difficulties for theoretical and numerical analysis. The Boltzmann equation, while possessing a wider applicability than hydrodynamic equations, requires significantly more computational resources due to the increased degrees of freedom in the model. The success of a hybrid fluid-kinetic flow solver for the study of multi-scale flows relies on accurate prediction of flow regimes. In this paper, we draw on binary classification in machine learning and propose the first neural network classifier to detect near-equilibrium and non-equilibrium flow regimes based on local flow conditions. Compared with classical semi-empirical criteria of continuum breakdown, the current method provides a data-driven alternative where the parameterized implicit function is trained by solutions of the Boltzmann equation. The ground-truth labels are derived rigorously from the deviation of particle distribution functions and the approximations based on the Chapman-Enskog ansatz. Therefore, no tunable parameter is needed in the criterion. Following the entropy closure of the Boltzmann moment system, a data generation strategy is developed to produce training and test sets. Numerical analysis shows its superiority over simulation-based samplings. A hybrid Boltzmann-Navier-Stokes flow solver is built correspondingly with adaptive partition of local flow regimes. Numerical experiments including one-dimensional Riemann problem, shear flow layer and hypersonic flow around circular cylinder are presented to validate the current scheme for simulating cross-scale and non-equilibrium flow physics. The quantitative comparison with a semi-empirical criterion and benchmark results demonstrates the capability of the current neural classifier to accurately predict continuum breakdown.
Recommendations
- Using neural networks to accelerate the solution of the Boltzmann equation
- Neural-network based collision operators for the Boltzmann equation
- RelaxNet: a structure-preserving neural network to approximate the Boltzmann collision operator
- Physics-informed neural networks for solving forward and inverse flow problems via the Boltzmann-BGK formulation
- DeepM\&Mnet for hypersonics: predicting the coupled flow and finite-rate chemistry behind a normal shock using neural-network approximation of operators
Cites work
- A hybrid continuum/particle approach for modeling subsonic, rarefied gas flows.
- A Model for Collision Processes in Gases. I. Small Amplitude Processes in Charged and Neutral One-Component Systems
- A stochastic kinetic scheme for multi-scale plasma transport with uncertainty quantification
- A unified gas-kinetic scheme for multiscale and multicomponent flow transport
- A velocity-space adaptive unified gas kinetic scheme for continuum and rarefied flows
- Convex Duality and Entropy-Based Moment Closures: Characterizing Degenerate Densities
- Data-driven discovery of governing equations for fluid dynamics based on molecular simulation
- Direct modeling for computational fluid dynamics. Construction and application of unified gas-kinetic schemes
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Domain of definition of Levermore's five-moment system
- Entropy-based moment closures for kinetic equations
- Fast algorithms for computing the Boltzmann collision operator
- General entropic approximations for canonical systems described by kinetic equations
- Generation of the Chapman-Enskog distribution
- Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations
- scientific article; zbMATH DE number 40939 (Why is no real title available?)
- scientific article; zbMATH DE number 1461212 (Why is no real title available?)
- MAXIMUM ENTROPY FOR REDUCED MOMENT PROBLEMS
- Maximum entropy moment systems and Galilean invariance
- Moment closure hierarchies for kinetic theories.
- Moment realizability and the validity of the Navier–Stokes equations for rarefied gas dynamics
- Physics-informed neural networks for high-speed flows
- Physics-informed neural networks for solving forward and inverse flow problems via the Boltzmann-BGK formulation
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Predicting failure of the continuum fluid equations in transitional hypersonic flows
- Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data
- SwitchNet: a neural Network Model for forward and inverse scattering problems
- The kinetic scheme for the full-Burnett equations
- The Stone-Weierstrass Theorem
- Using neural networks to accelerate the solution of the Boltzmann equation
- Variational training of neural network approximations of solution maps for physical models
Cited in
(4)- DeepStSNet: reconstructing the quantum state-resolved thermochemical nonequilibrium flowfield using deep neural operator learning with scarce data
- RelaxNet: a structure-preserving neural network to approximate the Boltzmann collision operator
- On the Invalidity of the Extended Navier-Stokes Equations to Compute Rarefied Gas Flows in a Cylinder Array
- Structure-preserving neural networks for the regularized entropy-based closure of a linear, kinetic, radiative transport equation
This page was built for publication: Predicting continuum breakdown with deep neural networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6107123)
