Abstract:High consumption of memory and computational effort is the major barrier to prevent the widespread use of the discrete velocity method (DVM) in the simulation of flows in all flow regimes. To overcome this drawback, an implicit DVM with a memory reduction technique for solving a steady discrete velocity Boltzmann equation (DVBE) is presented in this work. In the method, the distribution functions in the whole discrete velocity space do not need to be stored, and they are calculated from the macroscopic flow variables. As a result, its memory requirement is in the same order as the conventional Euler/Navier-Stokes solver. In the meantime, it is more efficient than the explicit DVM for the simulation of various flows. To make the method efficient for solving flow problems in all flow regimes, a prediction step is introduced to estimate the local equilibrium state of the DVBE. In the prediction step, the distribution function at the cell interface is calculated by the local solution of DVBE. For the flow simulation, when the cell size is less than the mean free path, the prediction step has almost no effect on the solution. However, when the cell size is much larger than the mean free path, the prediction step dominates the solution so as to provide reasonable results in such a flow regime. In addition, to further improve the computational efficiency of the developed scheme in the continuum flow regime, the implicit technique is also introduced into the prediction step. Numerical results showed that the proposed implicit scheme can provide reasonable results in all flow regimes and increase significantly the computational efficiency in the continuum flow regime as compared with the existing DVM solvers.

Flow Decomposition for Heat Equations with Memory

Observability for Heat Equations with Time-Dependent Analytic Memory

Controllability and Observability of a Heat Equation with Hyperbolic Memory Kernel

Numerical solution of the heat conduction problem with memory

Flow Decomposition Reveals Dynamical Structure of Markov Process

Dynamic Behavior of A Heat Equation with Memory

Machine Learning for the Flow Prediction of Fluids with Memory Effects on the Stress

A study of non-smooth convex flow decomposition

Analysis of wave propagation in the presence of a continuous line heat source under heat transfer with memory dependent derivatives

Memory effect on the thermoelastic responses of a porous half-space with a novel fractional heat conduction model

Flow field modulation.

Homogenization of degenerate coupled transport processes in porous media with memory terms

Memory-dependent derivative versus fractional derivative (II): Remodelling diffusion process

Decomposing force fields as flows on graphs reconstructed from stochastic trajectories

Flow updates for domain decomposition of entropic optimal transport

Adaptive Space-Time Domain Decomposition for Multiphase Flow in Porous Media with Bound Constraints

A note on flow polynomials of graphs

A Decomposition Theorem for Dynamic Flows

Memory behaviors of entropy production rates in heat conduction

An Implicit Scheme with Memory Reduction Technique for Steady State Solutions of DVBE in All Flow Regimes