Dynamic Smagorinsky-Lilly Model Research Articles

Verification and Validation for Large Eddy Simulation of Cavitating Flow Around a Projectile Near the Free Surface

AbstractVerification and validation (V&V) is a helpful tool for evaluating simulation errors, but its application in unsteady cavitating flow remains a challenging issue due to the difficulty in meeting the requirement of an asymptotic range. Hence, a new V&V approach for large eddy simulation (LES) is proposed. This approach offers a viable solution for the error estimation of simulation data that are unable to satisfy the asymptotic range. The simulation errors of cavitating flow around a projectile near the free surface are assessed using the new V&V method. The evident error values are primarily dispersed around the cavity region and free surface. The increasingly intense cavitating flow increases the error magnitudes. In addition, the modeling error magnitudes of the Dynamic Smagorinsky–Lilly model are substantially smaller than that of the Smagorinsky–Lilly model. The present V&V method can capture the decrease in the modeling errors due to model enhancements, further exhibiting its applicability in cavitating flow simulations. Moreover, the monitoring points where the simulation data are beyond the asymptotic range are primarily dispersed near the cavity region, and the number of such points grows as the cavitating flow intensifies. The simulation outcomes also suggest that the re-entrant jet and shedding cavity collapse are the chief sources of vorticity motions, which remarkably affect the simulation accuracy. The results of this study provide a valuable reference for V&V research.

Parametric study on flow characteristics of air vortex ring supply using large-eddy simulation

Air vortex-ring-based supply mode was proposed to deliver fresh air. The influence of air supply parameters on the entrainment and mixing of air vortex rings was investigated numerically. The air supply parameters, including the supply air velocity, formation time, and supply air temperature difference, were specified by theoretical analysis based on both the simple cylindrical-slug model and the application scenario. A large eddy simulation with a dynamic Smagorinsky–Lilly model was used to generate the data required for analysis. Sensitivity tests of the computational parameters and experimental validations were conducted to ensure the accuracy and applicability of the numerical method. The results showed that when the formation time was less than the formation number, increasing the formation time reduced the entrance gap width and consequently restricted the entrainment of the air vortex ring. When the formation time exceeded the formation number, the additional air appeared as a trailing wake and enhanced the diffusive entrainment. If the supply air velocity is too high, the entrainment and mixing process of the air diffused from the boundary of the vortex ring is accelerated. However, if the supply air velocity is too low, the air vortex ring collapses downstream, increasing the amount of pulsed air jet deposited behind the vortex ring. The non-isothermal pulsed jet caused more air from the pulsed jet to be deposited behind the vortex ring and accelerated the attenuation rate of the air vortex ring, indicating that the entrainment was enhanced.

104 Application of CFD Simulation for Development Cyclone Samplers

Abstract Aerosol sampling is one of the largest and growing application areas of small-scale cyclone separators. A series of limit values are already or will be reduced in the future and therefore there is a need for samplers operating at larger flow rates that can collect the required amount of particles for analysis at lower aerosol concentrations. Therefore, the main goal of this work was to develop and validate a Computational Fluid Dynamics (CFD) model that can be used for simulation and optimisation of cyclone samplers. For aerosol flow modelling the Euler-Lagrange method was applied where the continuous phase is treated in an Eulerian manner, whereas the dispersed phase is treated in a Lagrangian approach. For a proper simulation of the flow field, the Large Eddy Simulation (LES) technique was used. A dynamic Smagorinsky-Lilly model was applied to calculate subgrid-scale turbulence. The developed model was validated on two cyclone samplers, a 9.5 mm in diameter sampling cyclone of the HD design and a 23 mm in diameter FSP 10 cyclone. For that, the simulated performance characteristics, such as a cut-size, a slope of the penetration curve, and the pressure drop, were compared with the experimental measurements. The simulation results agree well with experimental data. The deviation in cut-size, the slope of the penetration curve, and the pressure drop between the experimental data and LES simulations of the HD cyclone do not exceed 7%, 18.4%, and 10.3% respectively. Regarding the FSP10 cyclone, the deviations in cyclone performance characteristics are even lower.

Noise characteristic analysis and sound sources identification for rod–airfoil interaction using different subgrid-scale models

Four subgrid-scale models based on large eddy simulation (LES), such as Smagorinsky–Lilly (SL), dynamic Smagorinsky–Lilly (DSL), wall-adapting local eddy-viscosity (WALE), and dynamic kinetic-energy transport (KET) were used and couple Ffowcs Williams–Hawkings equation to accurately analyze and identify the characteristics and position of the sound sources of rod–airfoil interaction. The results of four models were compared with experimental data. It was found that the DSL model was the optimal subgrid-scale model for the study of the interaction noise considering the calculation accuracy. Therefore, the DSL model was selected for analyzing and identifying the characteristics and location of the interaction noise source. During the calculation, solid and permeable data surfaces were used for acoustic integral surfaces. The results show that the impact of the quadrupole source is negligible at a low Mach number, and the dipole noise coming from the pressure fluctuations is dominant. Meanwhile, the dipole noise from the airfoil is louder than that from the rod; the leading edge of about 30% chord length of airfoil the is the main sound source of interference effect. Above results can provide guidance for research of blade-vortex interaction noise.

Calculation of External Vehicle Aerodynamic Noise Based on LES Subgrid Model

A dynamic Smagorinsky–Lilly model (DSLM) subgrid model on the basis of the Smagorinsky–Lilly subgrid model (SLM) was introduced in the OpenFOAM software. The flow field of the vehicle was simulated, and the pressure coefficient and sound pressure curve of the monitoring points were compared with the wind tunnel test results. The results show that the DSLM subgrid model with a wall function can achieve high simulation accuracy. The investigation of the flow field structure revealed an intermittent detachment of the turbulent vortex after the airflow passed through the rearview mirror, thereby resulting in a violent pressure pulsation on the side window around the rearview mirror. Airflow passed through the A-pillar, separated, and reattached on the upper side window, thereby producing aerodynamic noise. The research results can serve as a good reference for the simulation and test of aerodynamic noises outside the vehicle, and for the reduction of the aerodynamic noises of vehicles.

Large eddy simulation of pollutant dispersion in a naturally cross-ventilated model building: Comparison between sub-grid scale models

In the present article, the flow field and passive gaseous pollutant dispersion in a naturally cross-ventilated isolated single-zone model building have been investigated. The large eddy simulation (LES) approach has been applied together with three different sub-grid scale (SGS) models, namely, wall-adapting local eddy-viscosity (WALE), dynamic Smagorinsky-Lilly model (DSLM), and standard Smagorinsky-Lilly model (SSLM). The flow and passive scalar concentration fields have been compared with available experimental data. It is demonstrated that the three SGS models predict similar flow field. Also, it can be observed that the mean concentration field results obtained from the WALE sub-grid scale (SGS) model are in better agreement with the experimental data than those of DSLM and SSLM by about 8% and 5%, respectively. Moreover, the CPU time required for the computations by the WALE and SSLM models was almost 20% less than that of DSLM, making WALE a more suitable SGS model than the other two models for the prediction of flow and concentration fields in an isolated building. Furthermore, contributions of convective flux and turbulent diffusion flux to the pollutant transportation have been studied. It has been shown that although the convective flux is the main mechanism, the two fluxes have a significant influence on pollutant transportation and distribution.

A Method of Measuring Turbulent Flow Structures With Particle Image Velocimetry and Incorporating Into Boundary Conditions of Large Eddy Simulations

The setup of inlet conditions for a large eddy simulation (LES) is a complex and important problem. Normally, there are two methods to generate the inlet conditions for LES, i.e., synthesized turbulence methods and precursor simulation methods. This study presents a new method for determining inlet boundary conditions of LES using particle image velocimetry (PIV). LES shows sensitivity to inlet boundary conditions in the developing region, and this effect can even extend into the fully developed region of the flow. Two kinds of boundary conditions generated from PIV data, i.e., steady spatial distributed inlet (SSDI) and unsteady spatial distributed inlet (USDI), are studied. PIV provides valuable field measurement, but special care is needed to estimate turbulent kinetic energy and turbulent dissipation rate for SSDI. Correlation coefficients are used to analyze the autocorrelation of the PIV data. Different boundary conditions have different influences on LES, and their advantages and disadvantages for turbulence prediction and static pressure prediction are discussed in the paper. Two kinds of LES with different subgrid turbulence models are evaluated: namely dynamic Smagorinsky–Lilly model (Lilly model) and wall modeled large eddy simulation (WMLES model). The performances of these models for flow prediction in a square duct are presented. Furthermore, the LES results are compared with PIV measurement results and Reynolds-stress model (RSM) results at a downstream location for validation.

Experimental and LES investigation of premixed methane/air flame propagating in a tube with a thin obstacle

In this paper, an experimental and numerical investigation of premixed methane/air flame dynamics in a closed combustion vessel with a thin obstacle is described. In the experiment, high-speed video photography and a pressure transducer are used to study the flame shape changes and pressure dynamics. In the numerical simulation, four sub-grid scale viscosity models and three sub-grid scale combustion models are evaluated for their individual prediction compared with the experimental data. High-speed photographs show that the flame propagation process can be divided into five stages: spherical flame, finger-shaped flame, jet flame, mushroom-shaped flame and bidirectional propagation flame. Compared with the other sub-grid scale viscosity models and sub-grid scale combustion models, the dynamic Smagorinsky–Lilly model and the power-law flame wrinkling model are better able to predict the flame behaviour, respectively. Thus, coupling the dynamic Smagorinsky–Lilly model and the power-law flame wrinkling model, the numerical results demonstrate that flame shape change is a purely hydrodynamic phenomenon, and the mushroom-shaped flame and bidirectional propagation flame are the result of flame–vortex interaction. In addition, the transition from “corrugated flamelets” to “thin reaction zones” is observed in the simulation.

Large eddy simulation of a pitched blade impeller mixed vessel – Comparison with LDA measurements

This study deals with a large eddy simulation (LES) of the single-phase turbulent flow in a fully baffled mixed vessel stirred by a pitched six-blade impeller. The ANSYS FLUENT 15.7 commercial CFD programme was employed. The LES turbulent model is used with the dynamic Smagorinsky–Lilly model. The Werner–Wengle wall function is used to cope with the requirements of near the wall cell size of the LES model. The sliding mesh (SM) model was used for simulating the impeller movement. The results of the simulations are compared with experimental data obtained by previous LDA measurements of the radial profiles of the axial component of the ensemble averaged mean velocity in the impeller discharge flow. The volume flow rates and the flow rate criteria were calculated from these velocity profiles, and were compared with the flow rates determined from the horizontal vessel cross section from simulation.

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.

Effects of wind speed changes on wake instability of a wind turbine in a virtual wind tunnel using large eddy simulation

Large Eddy Simulation (LES) of the National Renewable Energy Laboratory (NREL) Phase VI wind turbine inside a virtual wind tunnel, with the same test section as that of NASA Ames 24.4m×36.6m, was carried out in order to analyze and better understand the wake instability and its breakdown behind the wind turbine. LES was performed using the commercial CFD software, ANSYS FLUENT, based on the dynamic Smagorinsky–Lilly model. The wind turbine was placed at a distance of two rotor diameters from the upstream boundary with a downstream domain of 20 rotor diameters in length. The results of the simulation were compared with the experimental data published by the NREL and a good agreement was found between the two. Furthermore, the average turbulence intensities from the LES were compared with a semi-empirical model and very good agreement was observed, except for the regions of on-going wake instability and vortex breakdown. It was observed that the wake behind the wind turbine consists of a system of intense and stable rotating helical vortices. These vortices persisted for some distance downstream of the wind turbine and finally become unstable producing a sinuous shape. The downstream distance at which wake instability and vortex breakdown occur, was observed to be a function of the upstream wind speed. For example, for an upstream wind speed of 7m/s, it was observed that the primary vortex structure became unstable at a downstream distance of four rotor diameters and complete breakdown occurred at approximately six rotor diameters. On the other hand, when the upstream wind speed was 15.1m/s, wake instability occurred at approximately 11 rotor diameters downstream of the wind turbine and complete breakdown was observed at 13 rotor diameters downstream of the wind turbine. Furthermore, it was observed that the turbulence intensity rapidly decreased during the process of wake instability and vortex breakdown; the location of the decrease is a function of the upstream wind speed. It is suggested that the distinction between the near and far wake can be identified as the average location between the start of the wake instability and the end of the process, at complete breakdown. Therefore the average location of this boundary is a function of the upstream wind speed. Hence for upstream wind speeds of 7m/s, 10m/s, 13.1m/s and 15.1m/s, the boundary between the near and far wake lies at five, seven, ten and twelve rotor diameters downstream respectively.

Large eddy simulation of the wind turbine wake characteristics in the numerical wind tunnel model

Large Eddy Simulation of NREL Phase VI wind turbine was performed in a virtual wind tunnel (24.4m×36.6m) in order to achieve a better understanding of the turbine wake characteristics. For this purpose, ANSYS-Fluent package was used to run the simulation using the dynamic Smagorinsky-Lilly model. For the purpose of validation, the pressure distribution at different span-wise sections along the turbine blade and the power produced by the wind turbine were compared with the published experimental results for the NREL phase VI rotor tested in the NASA wind tunnel with the same dimensions as in the model and a good agreement was found between the two. The airflow immediately behind the wind turbine was observed to be a system of intense and stable rotating helical vortices, which determined the dynamics of the far-wake. The system of vortices in the near-wake became unstable and broke down due to wake instability at a distance of five rotor diameters downstream of the wind turbine. This was defined as the boundary between the near- and far-wake regions. The collapsed spiral wake was found to spread in all directions in the far-wake resulting in the formation of the two pairs of counter-rotating vortices which caused the gradual increase of turbulence in these regions. The turbulence intensity in the wake was observed to increase immediately behind the turbine with a maximum of 12.12% at a distance of three rotor diameters downstream of the turbine, after which a gradual decrease in the turbulence intensity was observed in the near-wake regions due to wake instability. However, in the far-wake regions, due to counter-rotating vortices formed by the wake instability, the turbulence intensity showed a tendency to increase intensity. Finally the time-averaged wake velocities from the LES, with and without the blockage corrections, were compared with WAsP and a comparatively good agreement for the axial velocity predictions was observed in the far-wake.

Comparison of Reynolds averaged Navier-Stokes based simulation and large-eddy simulation for one isothermal swirling flow

The flow structure of one isothermal swirling case in the Sydney swirl flame database was studied using two numerical methods. Results from the Reynolds-averaged Navier-Stokes (RANS) approach and large eddy simulation (LES) were compared with experimental measurements. The simulations were applied in two different Cartesian grids which were investigated by a grid independence study for RANS and a post-estimator for LES. The RNG k-ɛ turbulence model was used in RANS and dynamic Smagorinsky-Lilly model was used as the sub-grid scale model in LES. A validation study and cross comparison of ensemble average and root mean square (RMS) results showed LES outperforms RANS statistic results. Flow field results indicated that both approaches could capture dominant flow structures, like vortex breakdown (VB), and precessing vortex core (PVC). Streamlines indicate that the formation mechanisms of VB deducted from the two methods were different. The vorticity field was also studied using a velocity gradient based method. This research gained in-depth understanding of isothermal swirling flow.

CFD Modelling of Liquid Homogenization in Stirred Tanks with One and Two Impellers Using Large Eddy Simulation

A prediction of liquid homogenization in stirred tanks using a CFD code FLUENT is presented. The study was performed for tanks agitated by one and two impellers on a centrally-located shaft. Two types of impellers were used: a six-blade 45° pitched blade turbine and a standard Rushton turbine. Different methods were employed for simulations of fluid flow in the stirred tanks—the multiple reference frames technique together with the standard k– ɛ turbulence model, the sliding mesh technique with the standard k– ɛ turbulence model, and the sliding mesh technique with the large eddy simulation (LES) model and a non-iterative time-advancement algorithm. The dynamic Smagorinsky–Lilly model using a locally calculated subgrid scale dynamic viscosity constant was used in the LES calculations. Tracer mass fractions in the tanks were recorded during time-dependent species transport simulations. The calculated flow field results—velocity profiles, power and pumping numbers were compared with the experimental results from literature. Time traces of normalized concentrations obtained from the simulations were compared with our experiments and the resulting mixing times were compared to literature correlations. The LES approach was the most time-consuming method; however, it described the real flow in stirred tanks better and we obtained more realistic courses of the liquid homogenization and the best agreement of the computed and experimental homogenization times of all the used models.