Dynamic Subgrid Scale Research Articles

Experimental and numerical investigation of a three-dimensional vertical-axis wind turbine with variable-pitch

A combined experimental and numerical investigation is carried out to study the performance of a micro vertical-axis wind turbine (VAWT) with variable-pitch. Three-dimensional numerical simulations are essentially employed, for the VAWT involves a low aspect ratio (AR) three straight blades with struts. The performance of the VAWT is experimentally measured using a wind tunnel, while large eddy simulation (LES) with dynamic smagorinsky subgrid scale (SGS) model is employed to help understand the associated flow structure. The effects of wind speed, turbulence intensity, airfoil shape, and strut mechanism with and without variable-pitch on the performance of the turbine are carefully assessed, both experimentally and numerically. The accuracy of the SGS model in predicting the laminar–turbulent transition is also examined.

Application of modified eddy dissipation concept with large eddy simulation for numerical investigation of internal combustion engines

Computational methodology of large eddy simulation (LES) is tested in a real internal combustion engine geometry. Premixed turbulent combustion and the behavior of flame propagation are studied and compared with different LES and combustion process models. KIVA-3V, an open source Fortran code, is modified to implement large-eddy simulation technique based on one-equation sub-grid scale (SGS) turbulent kinetic energy and Smagorinsky models. For combustion process, improved turbulent-mixing combustion models based on eddy-dissipation model (EDM) and modified eddy-dissipation concept (EDC) are implemented and studied in case of using LES. The models with modified combustion processes of EDM and EDC are validated using experimental data of premixed combustion of laboratory burner with application of converging optimum and desired combustion model constants. Also experimental in-cylinder pressure data are used to validate the numerical CFD results of engine geometry. The physics of flow structure, flame propagation, heat energy released and fuel mass consumption during combustion are studied and compared by LES models. It is observed that large-scale of unsteady turbulent flow motions are developed before spark ignition that makes different behavior of flame front shape in combustion process with different turbulent models. Also numerical simulations indicate that different LES turbulence models may not capture the same flame propagation structure, whereas LES dynamic SGS viscosity models in comparison with standard constant models, almost capture the same flame front shape and propagation structure in the combustion chamber.

Three-Dimensional Numerical Simulation of the Flow around Two Side-by-Side Cylinders of Different Diameters

This paper presents the calculation of hydrodynamic characteristics of two side-by-side cylinders of different diameters in three dimensional incompressible uniform cross flow by using Large-eddy simulation method based on dynamical Smagorinsky-Lilly sub-grid scale model. Solution of the three dimensional N-S equations were obtained by the finite volume method. The numerical simulation focused on investigating the characteristic of the pressure distribution (drag and lift force), vorticity field and turbulence Re=. Results shows that, the asymmetry of the time –averaged velocity distribution in the flow direction behind the two cylinders is very obvious; the frequency of eddy shedding of the small cylinder is about twice of the large one. The turbulence of cylinders is more obvious.

Dynamics of single, non-spherical ellipsoidal particles in a turbulent channel flow

Using disk, spherical and needle-like particles with equal equivalent volume diameters, the orientational dynamics of non-spherical particles is studied in a turbulent channel flow. An Eulerian–Lagrangian approach based on large eddy simulation with a dynamic sub-grid scale model is used to predict a fully developed gas–solid flow at a shear Reynolds number Reτ=300. Particle shape and orientation are accounted for by the coupling between Newton׳s law of translational motion and Euler׳s law of rotational motion, both in a Lagrangian framework. The particle shapes are simulated using the super-quadrics form, with the dynamically relevant parameters being the particle aspect ratio, equivalent volume diameter and response time. The translational and orientational behaviour of single particles initially released at three different locations in the wall-normal direction are monitored, with analysis showing a clear distinction between the behaviour of the different particle shapes. The results show that turbulent dispersion forces non-spherical particles to have a broad orientation distribution. The orientational states observed include periodic, steady rotation, tumbling, precessing and nutating. Velocity gradient, aspect ratio and particle inertia all have an effect on the alignment of the particle principal axis to the inertial axes. Unlike spherical particles, the disk and needle-like particles display a transition from one orientational state to another, especially when their initial position is in the near-wall region, with the frequency of this transition highest for the disk-like particle. Overall, this study leads to an improved understanding of the significance of shape on particle behaviour which is of relevance to many practical flows.

Performance of Transition-Sensitive Models in Predicting Flow Structures over Delta Wings

Laminar-turbulent transition may have considerable influence on the performance of moderate and high-speed aerodynamic vehicles. The two recently developed transition-sensitive models, SST-TR and , are employed to assess their abilities in resolving the complex underlying flow physics associated with flow structures over delta wings. In particular, their capabilities in capturing the aerodynamic characteristics of three-dimensional flow associated with vortex breakdown, shear-layer instabilities, and boundary-layer transition are assessed. These closures are employed to simulate vortex breakdown and near-wall flow structures over a delta wing planform with a moderate sweep angle of 50 deg at a Reynolds number of , where a combination of laminar, transitional, and turbulent flowfields coexist. Accuracy of these models in predicting boundary-layer transition adjacent to a delta wing surface with a high-sweep angle of 70 deg and Reynolds numbers of and is investigated. Large-eddy simulation with dynamic Smagorinsky subgrid scale model is also employed to assess its capability in predicting transition on delta wing planforms. All numerical simulations are compared with available experimental data found in the literature.

Numerical simulation of large dunes in meandering streams and rivers with in-stream rock structures

The evolution and migration of large dunes in a realistic intermediate-size experimental stream, the Saint Anthony Falls Laboratory (SAFL) Outdoor StreamLab (OSL), and two large-scale meandering rivers with in–stream rock structures are studied numerically using the SAFL Virtual StreamLab hydro-morphodynamic (VSL3D) model. Due to the challenges arising from mesh quality and large disparity in time-scales, coupled morpho- and hydro-dynamics simulations of bed forms has, for the most part, been restricted to sand wave amplitudes of few centimeters. In this work, we overcome such difficulties by employing the immersed boundary approach and a dual time–stepping technique of the VSL3D model [63]. The VSL3D employs the curvilinear immersed boundary (CURVIB) method along with a suspended sediment load module and is capable of simulating turbulent stratified flows coupled with bed morphodynamic evolution in realistic riverine environments with arbitrarily complex hydraulic structures. Turbulence is handled either via large-eddy simulation (LES) with the dynamic Smagorinski subgrid scale model or unsteady Reynolds-averaged Navier Stokes (URANS) equations closed with the k-ω turbulence model. Simulations in the intermediate-scale OSL channel, in which we also collected experimental morphodynamic data, show that LES can capture the evolution and migration of bed forms with characteristics that are in good agreement with experimental measurements. The URANS model, however, fails to excite the bed instability in the OSL channel but captures realistic dune evolution in the two large-scale meandering rivers. This finding is especially important as it demonstrates the potential of the VSL3D model as a powerful tool for simulating morphodynamic evolution under prototype conditions. To our knowledge, our work is the first attempt to simulate large-scale bed forms in waterways with an order of magnitude disparity in spatial scales, from the ∼2.7m wide OSL channel to the 27m wide rivers. Accordingly, the height of the simulated dunes ranges from ∼0.2m to 2.0m and the wavelength ranges from ∼0.1m to 50m for the OSL and large-scale rivers, respectively. For all cases the statistical properties of the simulated bed forms are shown to agree well with those of bed forms observed in nature.

Large eddy simulation of a fully developed non-isothermal turbulent channel flow

Purpose – The purpose of this paper is to numerically study heated channel flow using direct numerical simulation (DNS) and large eddy simulation (LES) method. Using different domain size and different grid resolution it is show that filtering procedure is influenced and may results in very different solutions. Design/methodology/approach – Turbulent non-isothermal fully developed channel flow has been investigated using LES. The filtered Navier-stokes and energy equations were numerically solved with dynamic subgrid scale (SGS) model, standard Smagorinsky model or without additional model for the turbulent SGS stress and heat flux required to close the governing equations. Findings – The numerical LES results in comparison with the DNS data demonstrate that the LES computations may not always offers a reliable prediction of non-isothermal turbulent flow in open channel. It has been found that, even though the models reproduces accurately results for the flow field the thermal field computed using LES do not necessary match the DNS results. Introducing SGS model for scalar do not always show large improvement. One of the reason is thickness of hydrodynamic and thermal boundary layer. In the cases when boundary layers are very different it is not easy optimally set up control volumes in the domain. Originality/value – This is one of the first instance in which a results of numerical computations for different grid resolution, different stretching, SGS model is employed for non-isothermal turbulent channel flow. It shows that in the cases when boundary layers hydrodynamic and thermal are very different it is hardly find optimal grid resolution or stretching

Impact of dynamic subgrid-scale modeling in variational multiscale large-eddy simulation of bluff-body flows

The effects of dynamic subgrid-scale (SGS) modeling are investigated in variational multiscale (VMS) large-eddy simulation (LES) simulations of bluff-body flows. The spatial discretization is based on a mixed finite-element/finite-volume formulation on unstructured grids. In the VMS approach used in this work, the separation between the largest and the smallest resolved scales is obtained through a variational projection operator and a finite-volume cell agglomeration. The dynamic and non-dynamic versions of Smagorinsky and wall-adapted local eddy-viscosity SGS models are used to account for the effects of the unresolved scales. In the VMS approach, these effects are only modeled in the smallest resolved scales. The VMS-LES and classical LES approaches, combined with the considered dynamic and non-dynamic SGS models, are applied to the simulation of the flow around a circular cylinder at Reynolds numbers 3,900 and 20,000 and to the flow around a square cylinder at Reynolds numbers 22,000 and 175,000.

On the Development of a Dynamic Non-linear Closure for Large-Eddy Simulation of the Atmospheric Boundary Layer

A dynamic procedure is developed to compute the model coefficients in the recently introduced modulated gradient models for both momentum and scalar fluxes. The magnitudes of the subgrid-scale (SGS) stress and the SGS flux are estimated using the local equilibrium hypothesis, and their structures (relative magnitude of each of the components) are given by the normalized gradient terms, which are derived from the Taylor expansion of the exact SGS stress/flux. Previously, the two model coefficients have been specified on the basis of theoretical arguments. Here, we develop a dynamic SGS procedure, wherein the model coefficients are computed dynamically according to the statistics of the resolved turbulence, rather than provided a priori or ad hoc. Results show that the two dynamically calculated coefficients have median values that are approximately constant throughout the turbulent atmospheric boundary layer (ABL), and their fluctuations follow a near log-normal distribution. These findings are consistent with the fact that, unlike eddy-viscosity/diffusivity models, modulated gradient models have been found to yield satisfactory results even with constant model coefficients. Results from large-eddy simulations of a neutral ABL and a stable ABL using the new closure show good agreement with reference results, including well-established theoretical predictions. For instance, the closure delivers the expected surface-layer similarity profiles and power-law scaling of the power spectra of velocity and scalar fluctuations. Further, the Lagrangian version of the model is tested in the neutral ABL case, and gives satisfactory results.

Large eddy simulation of dispersion around an isolated cubic building: evaluation of localized dynamic $$k_\mathrm{SGS}$$ k SGS -equation sub-grid scale model

In the present study, the prediction accuracy of a dynamic one-equation sub-grid scale model for the large eddy simulation of dispersion around an isolated cubic building is investigated. For this purpose, the localized dynamic \(k_\mathrm{SGS} \)-equation model (LDKM) is employed and the results are compared with the available experimental data and two other classic sub-grid scale models, namely, standard Smagorinsky–Lilly model (SSLM) and dynamic Smagorinsky–Lilly model (DSLM). It is shown that the three SGS models give results in good agreement with experiment. However, near the ground level of the leeward wall, dimensionless time-averaged concentration, \(\langle K\rangle \), profile is not quite similar to the experimental data. It is also demonstrated that the LDKM predicts the values of \(\langle K\rangle \) on the roof, leeward and side walls more acceptably than the SSLM and DSLM. Whereas, the streamwise elongation of time-averaged structures of the plume shape is more over-estimated with the LDKM than with the other two SGS models. In terms of numerical difficulty, the LDKM is found to be stable and computationally reasonable. In addition, it does not suffer from a flow dependent constant such as the Smagorinsky coefficient employed in the SSLM model.

Lattice Boltzmann Method for the Simulation of High Reynolds Number Flows

The lattice Boltzmann method (LBM) is considered as an alternative scheme to the standard Navier-Stokes approach. To simulate the high Reynolds number turbulence, several approaches based on LBM have been proposed. Among them, 5 approaches including the direct numerical simulation, dynamic subgrid scale model, inertial range consistent subgrid model, very large eddy simulation and entropic lattice Boltzmann method are discussed in detail. Features including improvements and shortcomings of each approach are presented. Whats more, application prospects of these approaches in high Reynolds number turbulence simulations are pointed out.

Lattice Boltzmann Method for the Simulation of High Reynolds Number Flows

The lattice Boltzmann method (LBM) is considered as an alternative scheme to the standard Navier-Stokes approach. To simulate the high Reynolds number turbulence, several approaches based on LBM have been proposed. Among them, 5 approaches including the direct numerical simulation, dynamic subgrid scale model, inertial range consistent subgrid model, very large eddy simulation and entropic lattice Boltzmann method are discussed in detail. Features including improvements and shortcomings of each approach are presented. Whats more, application prospects of these approaches in high Reynolds number turbulence simulations are pointed out.

Computation of transitional flow past a circular cylinder using multiblock lattice Boltzmann method with a dynamic subgrid scale model

Lattice Boltzmann method (LBM) is a kinetic based numerical scheme for the simulation of fluid flow. While the approach has attracted considerable attention during the last two decades, there is a need for systematic investigation of its applicability for complex canonical turbulent flow problems of engineering interest, where the nature of the numerical properties of the underlying scheme plays an important role for their accurate solution. In this paper, we discuss and evaluate a LBM based on a multiblock approach for efficient large eddy simulation of three-dimensional external flow past a circular cylinder in the transitional regime characterized by the presence of multiple scales. For enhanced numerical stability at higher Reynolds numbers, a multiple relaxation time formulation is considered. The effect of subgrid scales is represented by means of a Smagorinsky eddy-viscosity model, where the model coefficient is computed locally by means of a dynamic procedure, providing better representation of flow physics with reduced empiricism. Simulations are performed for a Reynolds number of 3900 based on the free stream velocity and cylinder diameter for which prior data is available for comparison. The presence of laminar boundary layer which separates into a pair of shear layers that evolve into turbulent wakes impose particular challenge for numerical methods for this condition. The relatively low numerical dissipation introduced by the inherently parallel and second-order accurate LBM is an important computational asset in this regard. Computations using five different grid levels, where the various blocks are suitably aligned to resolve multiscale flow features show that the structure of the recirculation region is well reproduced and the statistics of the mean flow and turbulent fluctuations are in satisfactory agreement with prior data.

An intercomparison of subgrid models for large‐eddy simulation of katabatic flows

AbstractWe present an intercomparison of three subgrid‐scale (SGS) models for large‐eddy simulation (LES) of katabatic flows. The SGS closures we study include the Smagorinsky formulation, a scale‐invariant dynamic model, and a scale‐dependent dynamic model. Downslope fluxes of mass and buoyancy deficit are highly sensitive to SGS closure choice. Due to strong shear and strong stratification, the dynamic models show a reduction of the Smagorinsky coefficient near the surface relative to the Smagorinsky model. Furthermore, results from both scale‐dependent dynamic models suggest that the assumptions of scale invariance in the dynamic model are violated. We present a second set of experiments which focus on grid resolution requirements of eddy‐resolving simulations of drainage flows. Downslope buoyancy and mass flux predictions produced by the scale‐dependent dynamic model are more robust with respect to grid resolution and grid anisotropy than the scale‐invariant dynamic and Smagorinsky SGS models. Predictions of vertically integrated downvalley mass and buoyancy fluxes are more sensitive to vertical grid resolution than horizontal grid resolution.

Large eddy simulation of flow around two side-by-side spheres

Large eddy simulation, using a dynamic Smagorinsky sub-grid scale, is used for the prediction of flow structures around two identical spheres fixed side-by-side at a subcritical flow regime with the Reynolds number equal to 5,000. The QUICK discretization method is applied to discretize the convection terms of the Navier-Stokes equation by means of the finite volume approach. This work focuses mainly on the wake structures downstream of the two interactive spheres located at three various dimensionless separation distances between spheres such as G/D = 1.50, 2.00, and 3.00. The obtained results revealed that the interaction between wakes affects the flow structures downstream of spheres. The rate of this interaction is strongly altered as a function of separation distance. On the other hand, some flow data such as mean drag and mean lift coefficients are affected due to the wake interactions. Finally, examination of the Reynolds stress variation along the different lateral axis, L/D, revealed that the nozzle effect does not play a significant role on the turbulence characteristics beyond the G/D = 3.00.

An Assessment of Dynamic Subgrid-Scale Sea-Surface Roughness Models

Covered by waves with various lengths, the mobile sea surface represents a great challenge to the large-eddy simulation (LES) of atmospheric boundary layer flow over the ocean surface. In this study, we report recent developments and tests of dynamic modeling approaches for the subgrid-scale (SGS) sea-surface roughness for LES. In the model, introduced originally in Yang et al. (J. Fluid Mech., in press, 2013), the SGS roughness is quantified by an integral of the SGS wave spectrum, \(\sigma_{\eta}^{\Delta}\), weighted based on the wind-wave kinematics, with an unknown model coefficient αw as pre-factor. The coefficient αw is determined dynamically based on the basic constraint that the total surface drag force must be independent of the LES filter scale. The weighted integral \(\sigma_{\eta}^{\Delta}\) represents the effective amplitude of the SGS waves, for which five candidate models are reviewed. Following the computational tests presented in Yang et al. (J. Fluid Mech., in press, 2013), in this study the performance of the dynamic SGS sea-surface roughness models is assessed by a theoretical approach, in which the roughness model is coupled with the critical-layer theory of wind–wave interaction. This theoretical approach mimics the averaged behavior of the LES. Meanwhile, its low computation cost allows the assessment of the model performance over a wide range of parameters. The test results indicate that the dynamic modeling approach can reliably model the roughness length of the SGS waves without ad-hoc prescription of the model parameter αw. Also, we confirm that to model \(\sigma_{\eta}^{\Delta}\), an expression based on the kinematics of wind–wave relative motion achieves the best performance among the five candidate models considered.

Large Eddy Simulation of a Polydisperse Ethanol Spray Flame

Ethanol is identified as an interesting alternative fuel. In this regards, the predictive capability of combustion Large Eddy Simulation approach coupled to Lagrangian droplet dynamic model to retrieve the turbulent droplet dispersion, droplet size distribution, spray evolution and combustion properties is investigated in this paper for an ethanol spray flame. Following the Eulerian-Lagrangian approach with a fully two way coupling, the Favre-filtered low Mach number Navier-Stokes equations are solved on structured grids with dynamic sub-grid scale models to describe the turbulent carrier gas phase. Droplets are injected in polydisperse manner and generated in time dependent boundary conditions. They evaporate to form an air-fuel mixture that yields spray flame. Part of the ethanol droplets evaporates within the prevaporization area before reaching the combustion zone, making the flame to burn in a partially premixed regime. The chemistry is described by a tabulated detailed chemistry based on the flamelet generated manifold approach. The fuel, ethanol, is modeled by a detailed reaction mechanism consisting of 56 species and 351 reversible reactions. The simulation results including excess gas temperature, droplet velocities and corresponding fluctuations, droplet mean diameters and spray volume flux at different distances from the exit plane show good agreement with experimental data. Analysis of combustion spray features allows gaining a deep insight into the two-phase flow process ongoing.

Computational and experimental investigation of scour past laboratory models of stream restoration rock structures

Local scour of the streambed around three models of stream restoration rock structures, including a rock vane, a cross vane, and a J-hook vane, is investigated via laboratory experiments and numerical simulations. In the experimental study, a physical model of each rock structure is constructed via an assembly of rocks and installed in a straight mobile sand bed flume. Continuous bed topography measurements provide insight into the time evolution of the scour patterns downstream of the structures and yield comprehensive data sets for validating the numerical simulations in terms of scour patterns, maximum scour depths, and bar migration dynamics. The numerical simulations are carried out using the coupled, hydro-morphodynamic Curvilinear Immersed Boundary (CURVIB) method of Khosronejad et al. (2011) [17]. The mobile channel bed and the individual rocks comprising a stream restoration structure are discretized with an unstructured triangular mesh and treated as sharp-interface immersed boundaries embedded in the background curvilinear mesh used to discretize the flow domain. For each case, simulations are carried out solving both the unsteady Reynolds-averaged Navier–Stokes (URANS) equations closed with the k–ω model and filtered Large-Eddy Simulation (LES) equations closed with the dynamic Smagorinski subgrid scale model. Both the URANS and LES models yield flow and scour patterns in reasonable agreement with the measurements with the LES results being consistently in better overall agreement with the measurements. To our knowledge, the present study is the first attempt to simulate local scour patterns around realistic model of stream restoration rock structures by taking into account and directly modeling their arbitrarily complex geometrical features.

Refined subgrid-scale model for large-eddy simulation of helical turbulence

A refined two-term helical subgrid-scale (SGS) stress model with respect to that suggested by Li et al. [Phys. Rev. E 74, 026310 (2006)] is designed for large-eddy simulation (LES) of helical turbulence. The model coefficients in the new model are verified a priori to be scale invariant in inertial range, which proves that our model is local in scale. A dynamic method based on minimizing the residual resolved energy and helicity dissipations is suggested to simultaneously evaluate the coefficients of the mixed SGS model as the simulation progresses. In addition, an SGS helicity dissipation (or helicity flux) constraint condition is proposed to optimize the mixed two-term model. Both techniques are first tested and validated in the LES of forced isotropic helical turbulence. The statistical results are analyzed and compared with those obtained from the dynamic Smagorinsky model, the traditional dynamic mixed model, and the direct numerical simulation. It is found that the introduction of this dynamic procedure can help overcome the drawback of the traditional dynamic method which can not capture the negative helicity fluxes and SGS dissipations. The probability density functions of the energy flux and the conditioned helicity flux and SGS stress demonstrate that the helicity flux constrained dynamic SGS model can effectively predict the real SGS helical effects on the resolved scales, such as backscatters of energy and helicity, accurate helicity dissipation rate, and so on. The present models are also applied to the simulation of freely decaying isotropic turbulence with no apparent improvement observed in comparison with the traditional SGS models. The underlying reasons for these issues are addressed in detail.

Large Eddy Simulation of Unstably Stratified Turbulent Flow over Urban-Like Building Arrays

Thermal instability induced by solar radiation is the most common condition of urban atmosphere in daytime. Compared to researches under neutral conditions, only a few numerical works studied the unstable urban boundary layer and the effect of buoyancy force is unclear. In this paper, unstably stratified turbulent boundary layer flow over three-dimensional urban-like building arrays with ground heating is simulated. Large eddy simulation is applied to capture main turbulence structures and the effect of buoyancy force on turbulence can be investigated. Lagrangian dynamic subgrid scale model is used for complex flow together with a wall function, taking into account the large pressure gradient near buildings. The numerical model and method are verified with the results measured in wind tunnel experiment. The simulated results satisfy well with the experiment in mean velocity and temperature, as well as turbulent intensities. Mean flow structure inside canopy layer varies with thermal instability, while no large secondary vortex is observed. Turbulent intensities are enhanced, as buoyancy force contributes to the production of turbulent kinetic energy.