A family of dynamic finite difference schemes for large-eddy simulation

A family of dynamic low-dispersive finite difference schemes for large-eddy simulation is developed. The dynamic schemes are constructed by combining Taylor series expansions on two different grid resolutions. The schemes are optimized dynamically during the simulation according to the flow physics and dispersion errors are minimized through the real-time adaption of the dynamic coefficient. In case of DNS-resolution, the dynamic schemes reduce to the standard Taylor-based finite difference schemes with formal asymptotic order of accuracy. When going to LES-resolution, the schemes seamlessly adapt to dispersion-relation preserving schemes. The schemes are tested for large-eddy simulation of Burgers’ equation and numerical errors are investigated as well as their interaction with the subgrid model. Very good results are obtained.     

Construction of explicit and implicit dynamic finite difference schemes and application to the large-eddy simulation of the Taylor–Green vortex

A general class of explicit and implicit dynamic finite difference schemes for large-eddy simulation is constructed, by combining Taylor series expansions on two different grid resolutions. After calibration for Re→∞, the dynamic finite difference schemes allow to minimize the dispersion errors during the calculation through the real-time adaption of a dynamic coefficient. In case of DNS resolution, these dynamic schemes reduce to Taylor-based finite difference schemes with formal asymptotic order of accuracy, whereas for LES resolution, the schemes adapt to Dispersion-Relation Preserving schemes. Both the explicit and implicit dynamic finite difference schemes are tested for the large-eddy simulation of the Taylor–Green vortex flow and numerical errors are investigated as well as their interaction with the dynamic Smagorinsky model and the multiscale Smagorinsky model. Very good results are obtained.     

A velocity-estimation subgrid model constrained by subgrid scale dissipation

Purely dissipative eddy-viscosity subgrid models have proven very successful in large-eddy simulations (LES) at moderate resolution. Simulations at coarse resolutions where the underlying assumption of small-scale universality is not valid, warrant more advanced models. However, non-eddy viscosity models are often unstable due to the lack of sufficient dissipation. This paper proposes a simple modeling approach which incorporates the dissipative nature of existing eddy viscosity models into more physically appealing non-eddy viscosity SGS models. The key idea is to impose the SGS dissipation of the eddy viscosity model as a constraint on the non-eddy viscosity model when determining the coefficients in the non-eddy viscosity model. We propose a new subgrid scale model (RSEM), which is based on estimation of the unresolved velocity field. RSEM is developed in physical space and does not require the use of finer grids to estimate the subgrid velocity field. The model coefficient is determined such that total SGS dissipation matches that from a target SGS model in the mean or least-squares sense. The dynamic Smagorinsky model is used to provide the target dissipation. Results are shown for LES of decaying isotropic turbulence and turbulent channel flow. For isotropic turbulence, RSEM displays some level of backward dissipation, while yielding as good results as the dynamic Smagorinsky model. For channel flow, the results from RSEM are better than those from the dynamic Smagorinsky model for both statistics and instantaneous flow structures.     

各向同性紊流能量衰减的大涡模拟

在多重网格驱动的,高效且得到充分验证的有限体积方法的基础上发展了可压缩流大涡模拟的方法.空间离散采用Jameson的中心格式附加二阶和四阶耗散的方法,时间推进则采用了双时间步长的方法.亚格子剪切应力张量和亚格子热通量用Smagorinsky模型进行模拟.通过对各向同性紊流能量衰减的模拟来验证本方法的准确性和高效性,模拟得到的能量谱和紊流动能衰减历程与过滤后的CBC实验数据吻合良好.     

On discretization errors and subgrid scale model implementations in large eddy simulations

We analyze the impact of discretization errors on the performance of the Smagorinsky model in large eddy simulations (LES). To avoid difficulties related to solid boundaries, we focus on decaying homogeneous turbulence. It is shown that two numerical implementations of the model in the same finite volume code lead to significantly different results in terms of kinetic energy decay, time evolutions of the viscous dissipation and kinetic energy spectra. In comparison with spectral LES results, excellent predictions are however obtained with a novel formulation of the model derived from the discrete Navier–Stokes equations. We also highlight the effect of discretization errors on the measurement of physical quantities that involve scales close to the grid resolution.     

Comparison of Several Spatial Discretizations for the Navier–Stokes Equations

Grid convergence studies for subsonic and transonic flows over airfoils are presented in order to compare the accuracy of several spatial discretizations for the compressible Navier–Stokes equations. The discretizations include the following schemes for the inviscid fluxes: (1) second-order-accurate centered differences with third-order matrix numerical dissipation, (2) the second-order convective upstream split pressure scheme (CUSP), (3) third-order upwind-biased differencing with Roe's flux-difference splitting, and (4) fourth-order centered differences with third-order matrix numerical dissipation. The first three are combined with second-order differencing for the grid metrics and viscous terms. The fourth discretization uses fourth-order differencing for the grid metrics and viscous terms, as well as higher-order approximations near boundaries and for the numerical integration used to calculate forces and moments. The results indicate that the discretization using higher-order approximations for all terms is substantially more accurate than the others, producing less than two percent numerical error in lift and drag components on grids with less than 13,000 nodes for subsonic cases and less than 18,000 nodes for transonic cases. Since the cost per grid node of all of the discretizations studied is comparable, the higher-order discretization produces solutions of a given accuracy much more efficiently than the others.     

Multiscale three-point velocity increment correlation in turbulent flows

The three-point velocity increment correlation function is proposed to represent the multiscale correlations in turbulent flows. The inertial–inertial correlation and the inertial–dissipative correlation are discussed due to their endogenetic properties in turbulence and their roles in large-eddy simulation. The zero-correlation points are then emphasized as equilibrium points between them. The credibility of this theoretical result is numerically verified in both isotropic and anisotropic flows. Results imply the universality of this zero-correlation scaling in different turbulent flows. This work is expected to be a dependable theoretical base for creating multiscale subgrid models in large-eddy simulation.     

狭窄通道湍流纵向涡强化换热实验和数值研究

本文采用实验的方法,研究了单面加热矩形狭窄通道内,翼片型纵向涡发生器对流动换热的强化作用.在此基础上,应用大涡模拟的方法对通道内的瞬态流场及其对壁面对流换热的影响进行了研究,并将数值模拟与实验进行了比较.结果表明,通过添加翼片可以在流动中产生涡流,强化壁面边界层与流体的物质和能量交换,并验证了将大涡模拟应用于纵向涡强化换热研究的可行性.     

Adjoint and defect error bounding and correction for functional estimates

We present two error estimation approaches for bounding or correcting the error in functional estimates such as lift or drag. Adjoint methods quantify the error in a particular output functional that results from residual errors in approximating the solution to the partial differential equation. Defect methods can be used to bound or reduce the error in the entire solution, with corresponding improvements to functional estimates. Both approaches rely on smooth solution reconstructions and may be used separately or in combination to obtain highly accurate solutions with asymptotically sharp error bounds. The adjoint theory is presented for both smooth and shocked problems; numerical experiments confirm fourth-order error estimates for a pressure integral of shocked quasi-1D Euler flow. By employing defect and adjoint methods together and accounting for errors in approximating the geometry, it is possible to obtain functional estimates that exceed the order of accuracy of the discretization process and the reconstruction approach. Superconvergent drag estimates are obtained for subsonic Euler flow over a lifting airfoil.     

Comparisons of different implementations of turbulence modelling in lattice Boltzmann method

In this paper, we present an alternative approach for the turbulence modelling in the single-relaxation-time lattice Boltzmann method (LBM) framework by treating the turbulence term as an extra forcing term, in addition to the traditional approach of modifying the relaxation time. We compare these two different approaches and their mixture in large-eddy simulation (LES) of three-dimensional decaying isotropic homogenous turbulence using the Smagorinsky model and the mixed similarity model. When the LES was conducted using the Smagorinsky model, where the Boussinesq eddy-viscosity approximation is adopted, the results showed that these three different implementations are equivalent. However, when the mixed similarity model is adopted, which is beyond the Boussinesq eddy-viscosity approximation, our results showed that an equivalent eddy-viscosity will lead to errors, while the forcing approach is more straightforward and accurate. This provides an alternative and more general framework of simulation of turbulence with models in LBM, especially when the Boussinesq eddy-viscosity approximation is invalid.     

An Asymptotic-Preserving all-speed scheme for the Euler and Navier–Stokes equations

We present an Asymptotic-Preserving ‘all-speed’ scheme for the simulation of compressible flows valid at all Mach-numbers ranging from very small to order unity. The scheme is based on a semi-implicit discretization which treats the acoustic part implicitly and the convective and diffusive parts explicitly. This discretization, which is the key to the Asymptotic-Preserving property, provides a consistent approximation of both the hyperbolic compressible regime and the elliptic incompressible regime. The divergence-free condition on the velocity in the incompressible regime is respected, and an the pressure is computed via an elliptic equation resulting from a suitable combination of the momentum and energy equations. The implicit treatment of the acoustic part allows the time-step to be independent of the Mach number. The scheme is conservative and applies to steady or unsteady flows and to general equations of state. One and two-dimensional numerical results provide a validation of the Asymptotic-Preserving ‘all-speed’ properties.     

A numerical source of small-scale number-density fluctuations in Eulerian–Lagrangian simulations of multiphase flows

Eulerian–Lagrangian simulations of multiphase flow are known to suffer from two errors that can introduce small-scale fluctuations in the number-density of the dispersed phase. These errors can be reduced by increasing the number of particles in the simulation. Here, we present results to demonstrate that a third error exists that can also generate small-scale number-density fluctuations. In contrast to the two known errors, this error cannot be lowered by increasing the number of particles. Analysis shows that this error is caused by spatial variation at the subgrid scale in the interpolation error of the fluid velocity to the particle location. If the particle velocity divergence is zero, the particle concentration error remains at the subgrid scale. However, if particles preferentially accumulate either due to their inertia or due to divergence of the underlying fluid-velocity field, this error manifests as number-density fluctuations on the grid scale. The only mechanism of reducing these errors is through higher-order accurate interpolation. By studying two model problems, estimates for the errors are derived. These estimates are shown to be quite accurate for simulations of shock and expansion waves interacting with particles.     

格子Boltzmann亚格子模型的研究

为了将格子Boltzmann法应用于大雷诺数流动的模拟,本文将Smagorinsky亚格子模型和LBGK模型相结合,并对该亚格子LBM模型进行了研究。利用该亚格子LBM模型,对二维顶盖驱动流进行了模拟,得到了若干大雷诺数下流线图和方腔中心线上无量纲速度分布。计算结果与基准解进行比较,两者相互吻合。     

Development of an algebraic wall heat transfer model for LES in IC engines using DNS data

This paper presents a novel algebraic wall-modeled large-eddy simulation (WMLES) approach for wall heat transfer (WHT) in internal combustion engines (ICE) using recent high-fidelity simulation data from two real ICE and a flame-wall interaction (FWI) setup. The model formulation is based on intuitive arguments rather than simplified forms of near-wall governing equations. Input information from the two wall-normal wall-adjacent nodes is used, facilitating the interpretation of the local state of the thermal boundary layer (TBL). With filtered direct numerical simulation (DNS) data (i.e. a ‘perfect’ LES), model performance is evaluated locally at different near-wall resolutions. The proposed model is compared to a widely used wall function (WF) approach and to a data-driven model.     

理性亚格子建模的实践与反思

亚格子建模是大涡模拟中的核心问题之一.理性亚格子建模的初衷是为尽量去除建模过程中的经验和唯象的成分,让建模过程中的每一步都在物理和数学上有迹可循.根据理性亚格子建模的理论已经发展出一系列不依赖经验常数和唯象模型的亚格子模型,但在实际问题中的效果却仍有不足,从而引发了近年来对亚格子建模思想本身的反思.本文回顾已有的研究和新的进展,目标是对未来的大涡模拟给出理性且自洽的理论指导.     

Simulation of laminar–turbulent transition in compressible Taylor–Green flow basing on quasi-gas dynamic equations

We report the results of numerical simulation of laminar–turbulent transition in the Taylor–Green vortex for viscous compressible gas flow basing on quasi-gas-dynamic (QGD) equations. Here the QGD system is obtained by a temporal averaging of the Navier–Stokes equations. The additional dissipative terms in QGD system serve to model the effects of the unresolved subgrid scales. Comparison with direct numerical simulation and large eddy simulation reference data demonstrates that QGD numerical algorithm provides a uniform and adequate simulation of both laminar and turbulent evolution of the vortex for Reynolds numbers from 100 up to 5000, including transition.     

壁面在展向作周期运动的槽道湍流的大涡模拟

分别采用3种亚格子模式:传统的Smagorinsky模式、动力Smagorinsky模式和Cui(2004)基于Kolmogorov方程所提出的新模式,对壁面在展向作周期运动的槽道湍流进行了大涡模拟,以考察这3种模式对平均运动为三维、非定常的湍流流动的模拟能力.通过对湍流基本统计量的分析,发现动力模式和新模式都可以较好地预测这种三维非定常的湍流流动;对相位平均的湍流统计量,动力模式的结果略优于新模式;传统的Smagorinsky模式对这种流动的预测结果是最差的.     

A-priori testing of alpha regularisation models as subgrid-scale closures for large-eddy simulations

Alpha-type regularisation models provide theoretically attractive subgrid-scale closure approximations for large-eddy simulations of turbulent flow. We adopt the a-priori testing strategy to study three different alpha regularisation models, namely the Navier–Stokes-α model, the Leray-α model, and the Clark-α model. Specifically, we use high-resolution direct numerical simulation data of homogeneous isotropic turbulence to compute the mean subgrid-scale dissipation, the spatial distribution of the subgrid-scale dissipation, and the spatial distribution of elements of the subgrid-scale stress tensor. This is done for different filter parameters and different large-eddy simulation grid resolutions. Predictions of the three regularisation models are compared to the exact values of the subgrid-scale stress tensor, as defined in the filtered Navier–Stokes equations. The potential of the three regularisation models to provide good approximations is quantified using spatial correlation coefficients. Whereas the Clark-α model exhibits the highest spatial correlation coefficients for the subgrid-scale dissipation and the subgrid-scale stress tensor elements, the Leray-α model provides lower correlation coefficients, and the Navier–Stokes-α model exhibits the lowest correlation coefficients of the three models. Our results indicate the presence of an optimal choice of the filter parameter α depending on the large-eddy simulation grid resolution.