Lattice Boltzmann Method with regularized non-equilibrium distribution functions

Jonas Latt,Bastien Chopard

DOI: https://doi.org/10.48550/arXiv.physics/0506157

2005-06-20

Abstract:A new lattice Boltzmann (LB) model is introduced, based on a regularization of the pre-collision distribution functions in terms of the local density, velocity, and momentum flux tensor. The model dramatically improves the precision and numerical stability for the simulation of fluid flows by LB methods. This claim is supported by simulation results of some 2D and 3D flows.

Fluid Dynamics,Computational Physics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: when using the Lattice Boltzmann Method (LB) for fluid flow simulation, how to improve numerical stability and accuracy. Specifically, the LB method is prone to numerical instability at high Reynolds numbers, resulting in inaccurate simulation results or failure to converge.

To solve this problem, the author proposes a new model based on the regularized non - equilibrium distribution function. This model significantly improves the accuracy and stability of the LB method in simulating fluid flow by modifying the distribution function before collision to make it closer to the theoretically expected non - equilibrium part.

### Main improvement points

1. **Regularized non - equilibrium distribution function**:
By recalculating the non - equilibrium distribution function \( f^{\text{neq}} \), it is made to conform to the results of the first - order multi - scale Chapman - Enskog expansion:
\[
f_i^{(1)}=\frac{t_i}{2c_s^4}Q_{i\alpha\beta}\Pi_{\alpha\beta}^{\text{neq}}
\]
where \( Q_{i\alpha\beta}=v_{i\alpha}v_{i\beta}-c_s^2\delta_{\alpha\beta} \), and \( \Pi_{\alpha\beta}^{\text{neq}} \) is the non - equilibrium part of the momentum flux tensor.

2. **Simplified BGK collision step**:
In the new collision step, the regularized non - equilibrium distribution function is used to update the distribution function:
\[
f_i^{\text{out}}=f_i^{\text{eq}}+(1 - \omega)f_i^{(1)}
\]
This can ensure mass conservation and momentum conservation while improving numerical stability.

### Experimental verification

The author verifies the effectiveness of the new model through several two - dimensional and three - dimensional numerical experiments of fluid flow:

- **Kovasznay flow**: At different grid resolutions, the relative error of the regularized model is significantly lower than that of the traditional BGK model, and no numerical instability occurs on small grids.
- **Two - dimensional cavity flow**: For a given Reynolds number, the maximum stable Reynolds number of the regularized model is much higher than that of the BGK model, indicating that it has higher numerical stability.
- **Three - dimensional turbulent flow**: At high Reynolds numbers, the BGK model shows numerical instability, while the regularized model remains stable, and its statistical characteristics are consistent with the theoretical predictions of fluid turbulence.

In summary, the new model proposed in this paper significantly improves the accuracy and stability of the LB method in fluid flow simulation through regularizing the non - equilibrium distribution function, especially in applications at high Reynolds numbers and under complex boundary conditions.