Bernard J. Geurts,Darryl D. Holm
Abstract:We consider so-called Leray regularization of the convective contributions. This gives rise to a subgrid parameterization which involves both explicit filtering and (approximate) inversion. The Leray model also arises from the alpha-modeling strategy derived via Kelvin's circulation theorem. We study the dynamics associated with the Leray model in a turbulent mixing layer and compare predictions with filtered DNS results and findings due to dynamic (mixed) models. In particular, the kinetic energy, momentum thickness and energy-spectra are analyzed, establishing favorable performance of the Leray model and robustness at arbitrarily high Reynolds number. This is unique for a similarity-type model that does not contain an explicit eddy-viscosity term.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the high - resolution computing challenge in turbulence simulation, especially how to accurately capture the main features of turbulence with limited computing resources. Specifically, the paper explores the application of the Leray regularization method in simulating turbulent shear layers, hoping that through this method, the computational complexity can be effectively reduced while maintaining an accurate description of the main features of the flow.
### Problem Background
Although direct numerical simulation (DNS) can in principle accurately solve the Navier - Stokes equations, it is greatly limited in practical applications due to its extremely high requirements for spatial resolution. Therefore, researchers seek alternative methods, such as large - eddy simulation (LES), which simplifies the flow characteristics through spatial filtering, thereby reducing the computational complexity. However, this method introduces the "subgrid closure problem", that is, how to describe the nonlinear effects after filtering.
### Leray Regularization Method
Leray regularization is a method for dealing with nonlinear terms. It reduces the influence of nonlinear effects by smoothing the convection terms. Specifically, the Leray model replaces the convection term \(u_j\partial_j u_i\) in the original Navier - Stokes equations with \(\bar{u}_j\partial_j u_i\), where \(\bar{u}\) is the filtered velocity field. This makes the nonlinear effects regulated by the smoothing operation, thereby reducing the computational complexity.
### Research Content
1. **Basic Leray Formula**: The paper derives the basic formula of the Leray model and introduces how to implement filtering and inverse - filtering operations through the discrete Fourier transform.
2. **Turbulent Mixing Layer Simulation**: The simulation effects of the Leray model on turbulent mixing layers at different Reynolds numbers are studied and compared with DNS data and other dynamic models.
3. **Performance Evaluation**: Key parameters such as kinetic energy, momentum thickness and energy spectrum are analyzed to verify the robustness and accuracy of the Leray model at different Reynolds numbers.
### Main Findings
- **Kinetic Energy and Momentum Thickness**: The Leray model can well predict the momentum thickness and performs excellently in the forward and backward energy transfer of kinetic energy.
- **Robustness at High Reynolds Numbers**: Even at extremely high Reynolds numbers, the Leray model can still stably capture the flow characteristics, showing good grid independence.
- **Computational Efficiency**: Compared with the dynamic mixing model, the Leray model has lower computational costs and does not need to introduce additional empirical parameters.
### Conclusion
The Leray model performs excellently in simulating turbulent shear layers. It can accurately capture the main features of the flow while ensuring computational efficiency, especially outstanding under high - Reynolds - number conditions. This method provides a potentially new tool for future large - eddy simulations.