Baseline Reynolds Stress Model Research Articles

On the application of statistical turbulence models to the simulation of airflow inside a car cabin

Computational fluid dynamics simulations of airflow inside a full-scale passenger car cabin are performed using the Reynolds averaged Navier–Stokes equations. The performance of a range of turbulence models is examined by reference to experimental results of the streamwise mean velocity and turbulence intensity profiles, obtained using the hot-wire anemometry technique at different locations inside the car cabin. The models include three linear eddy-viscosity-based variants, namely, the realizable k–ε, the renormalization group k–ε, and the shear-stress transport k–ω models. The baseline Reynolds stress model (BSL-RSM), a second-moment-closure variant, and an Explicit Algebraic Reynolds Stress Model (BSL-EARSM) are also investigated. Visualization of velocity vectors and streamlines in different longitudinal planes shows a similar airflow pattern. The flow topology is mainly characterized by jet flows developing from the dashboard air vents and extending to the back-seats compartment resulting in a large vortex structure. Additionally, a comparison between numerical and experimental results shows a relatively good agreement of the mean velocity profiles. However, all models exhibit some limitations in predicting the correct level of turbulence intensity. Moreover, the realizability of the modeled Reynolds stresses and the structure of turbulence are analyzed based on the anisotropy invariant mapping approach. All models reveal a few amounts of non-realizable solutions. The linear eddy-viscosity-based models return a prevailing isotropic turbulence state, while the BSL-RSM and the BSL-EARSM models display pronounced anisotropic turbulence states.

Calibration of RANS model constant based on data assimilation and accurate simulation of separated flow

To improve the prediction accuracy of separated flow based on the Reynolds Averaged Navier–Stokes model, the model constants of the baseline Reynolds stress model are calibrated by the ensemble Kalman filter data assimilation method. The separated flow in a diffuser is taken as the object, and the wall pressure coefficients of the diffuser are used as the driving data. The results show that the method that recalibrates the model constants based on data assimilation is easy to implement and is an effective method. The wall pressure coefficients and the separation regions of the diffuser predicted by the baseline Reynolds stress model with the default model constants deviate greatly from the experimental observations. By recalibrating the model constants, the prediction accuracy of separated flow based on the baseline Reynolds stress model is improved. This provides an idea for the accurate simulation of separated flow based on the Reynolds Averaged Navier–Stokes model in engineering applications.

Computational modelling of turbulent Taylor–Couette flow for bearing chamber applications: A comparison of unsteady Reynolds-averaged Navier–Stokes models

The capability to accurately model fluid flow within rotating Taylor–Couette systems has a primary role in informing computational investigations of rotating machinery. There is considerable uncertainty regarding selection of modelling approach, including a suitable turbulence model, that can accurately resolve turbulence within such complex flows while remaining computationally feasible for industrially relevant applications. This paper presents a numerical comparison of axisymmetric and three-dimensional unsteady Reynolds-averaged Navier–Stokes (URANS) turbulence models within ANSYS Fluent. The CFD geometries are representative of ones for which there are published experimental measurements. For the Taylor–Couette study, investigation into inner cylinder start-up procedure, based on previous published findings, confirmed that the final state of the flow is highly dependent on the initial conditions and acceleration rate. Once Taylor vortices form and stabilise, they are not disrupted by small steps in inner cylinder speed, allowing computationally efficient accelerations. Investigations into applying rotational periodicity were unsuccessful, resulting in a significantly reduced core velocity. Axisymmetric predictions provided reasonable agreement with experimental data only at low rotation rates. A good prediction of the velocity flow field was obtained for three-dimensional simulations of the full 360° domain with differences of less than 5% for radial velocities. Among the URANS models, the standard k-ω model and baseline Reynolds stress model (BSL-RSM) provided the closest agreement to published experimental data. In the paper, the developed Taylor–Couette turbulence modelling methodology is extended to a bearing chamber geometry. Analysis of the secondary vortex flow field is compared both qualitatively and quantitatively to published bearing chamber experimental measurements. Overall, whilst a good agreement is still found using the standard k-ω turbulence model, discrepancies arise with the BSL-RSM. However, for this more complex bearing chamber environment compared to a Taylor–Couette flow, the shear stress transport k-ω turbulence model provided the closest agreement and is recommended for future bearing chamber modelling.

RANS-BASED CFD PREDICTION OF THE HYDRODYNAMIC COEFFICIENTS OF DARPA SUBOFF GEOMETRY IN STRAIGHT-LINE AND ROTATING ARM MANOEUVRES

Computational Fluid Dynamics (CFD) simulations using Reynolds Averaged Navier-Stokes (RANS) equations are increasingly adopted as an analysis tool to predict the hydrodynamic coefficients of underwater vehicles. These simulations have shown to offer both a high degree of accuracy comparable to experimental methods and a greatly reduced computational cost compared to Large Eddy Simulation (LES) and Direct Numerical Simulation (DNS). However, one of the major challenges faced with CFD simulations is that the results can vary greatly depending on the numerical model settings. This paper uses the DARPA SUBOFF hull form undergoing straight-line and rotating arm manoeuvres at different drift angles to analyse the hydrodynamic forces and moments on the vehicle against experimental data, showing that the selection of the boundary conditions and turbulence models, and the quality of the mesh model can have a considerable and independent effect on the computational results. Comparison between the Baseline Reynolds Stress Model (BSLRSM) and Shear Stress Transport with Curvature Correction (SSTCC) were carried out for both manoeuvres, showing that with a sufficiently fine mesh, appropriate mesh treatment, and simulation conditions matching the experiments; the BSLRSM predictions offer good agreement with experimental measurements, while the SSTCC predictions are agreeable with the longitudinal force but fall outside the experimental uncertainty for both the lateral force and yawing moment.

Modeling Surface Riblets Skin Friction Reduction Effect with the Use of Computational Fluid Dynamics

Towards the minimization of fuel consumption and aicraft emissions reduction, many disruptive technologies have been proposed in aviation industry regarding the improvement of aircraft aerodynamic performance. Among others, these technologies involve innovative passive and active flow control methodologies and techniques. A passive flow control methodology which is studied in the current work is the implementation of riblets, which are inspired by shark skin morphology, on a NACA 0012 airfoil. The main purpose of riblets is to alter the boundary layer characteristics near the wall region in such a way that the total skin friction decreases, resulting in an overall drag reduction of the aircraft with a straightforward impact on fuel consumption and thus emissions. The riblets are implemented in the Reynolds Averaged Navier Stokes equations with appropriate source terms in the turbulent dissipation transport equation. Two turbulence models are used, the k-? SST linear eddy viscosity model and the Baseline Reynolds Stress model which also uses the transport equation of the specific rate of turbulence dissipation (?). The computational results are compared with available experimental data of NACA 0012 with surface riblets attached. Drag and skin friction coefficients are presented for various angles of attack and the maximum potential benefit from riblet use is evaluated regarding aerodynamic performance and consequently reduction of emissions and fuel consumption.

The influence of turbulence model and two and three-dimensional domain selection on the simulated performance characteristics of vertical axis tidal turbines

The influence of Computational Fluid Dynamics (CFD) modeling techniques on the accuracy of fixed pitch vertical axis turbine power output predictions was investigated. Using Two-Dimensional (2D) and Three-Dimensional (3D) models, as well as the Baseline-Reynolds Stress Model (BSL-RSM) and the k-ω Shear Stress Transport (k-ω SST) model in its fully turbulent and laminar-to-turbulent formulation, differences in power output modeling accuracy were evaluated against experimental results from literature. The highest correlation was found using a 3D domain model that fully resolved the boundary layer combined with the k-ω SST laminar-to-turbulent model. The turbulent 3D fully resolved boundary layer k-ω SST model also accurately predicted power output for most rotational rates, at a significantly reduced computational cost when compared to its laminar-to-turbulent formulation. The 3D fully resolved BSL-RSM model and 3D wall function boundary layer k-ω SST model were found to poorly simulate power output. Poor output predictions were also obtained using 2D domain k-ω SST models, as they were unable to account for blade tip and strut effects. The authors suggest that 3D domain fully turbulent k-ω SST models with fully resolved boundary layer meshes are used for predicting turbine power output given their accuracy and computational efficiency.

Numerical analysis on supercritical water heat transfer in a 2 × 2 rod bundle

CFD analysis is carried out for supercritical water (SCW) heat transfer in a 2×2 rod bundle with three turbulence models, i.e. the SSG, the ω Reynolds stress model (ω-RSM) and the baseline Reynolds stress model (BSL-RSM). The CFD calculation results show that the models predict similar cross-flow pattern between the adjacent subchannels. However, there is apparent discrepancy in the predicted turbulent mixing. Due to the strong cross-flow, the bulk parameters predicted by the models agree well. Comparison between the CFD results and the experiment data reveals that the SSG underpredicts the wall temperature, while the ω-RSM and BSL-RSM tend to overpredict the wall temperature remarkably in the region where the buoyancy effect is dominant near the wall. Such difference can be mainly attributed to the modeling approach near the wall. It may be concluded that the appropriate modeling of buoyancy effect is the key factor to improve the prediction when the wall region is resolved with very fine mesh. The bulk flow parameter obtained in the CFD calculation which cannot be easily measured in the experiment are used to evaluate the SCW heat transfer correlation. An uncertainty analysis shows that the uncertainty caused by the selection of turbulence models is negligible for the correlation evaluation.

Rans-Based CFD Prediction of the Hydrodynamic Coefficients of Darpa Suboff Geometry in Straight-Line and Rotating Arm Manoeuvres

"Computational Fluid Dynamics (CFD) simulations using Reynolds Averaged Navier-Stokes (RANS) equations are increasingly adopted as an analysis tool to predict the hydrodynamic coefficients of underwater vehicles. These simulations have shown to offer both a high degree of accuracy comparable to experimental methods and a greatly reduced computational cost compared to Large Eddy Simulation (LES) and Direct Numerical Simulation (DNS). However, one of the major challenges faced with CFD simulations is that the results can vary greatly depending on the numerical model settings. This paper uses the DARPA SUBOFF hull form undergoing straight-line and rotating arm manoeuvres at different drift angles to analyse the hydrodynamic forces and moments on the vehicle against experimental data, showing that the selection of the boundary conditions and turbulence models, and the quality of the mesh model can have a considerable and independent effect on the computational results. Comparison between the Baseline Reynolds Stress Model (BSLRSM) and Shear Stress Transport with Curvature Correction (SSTCC) were carried out for both manoeuvres, showing that with a sufficiently fine mesh, appropriate mesh treatment, and simulation conditions matching the experiments; the BSLRSM predictions offer good agreement with experimental measurements, while the SSTCC predictions are agreeable with the longitudinal force but fall outside the experimental uncertainty for both the lateral force and yawing moment."

Laser Doppler measurement and CFD validation in 3 × 3 bundle flow

The five-beam three-component laser Doppler system is operated in the three-beam two-component mode to measure the 3×3 bundle flow with simple grid spacer. Experiment has been conducted at Re=15,200 and 29,900. According to the experiment result, the root mean square (RMS) of axial velocity fluctuation shows large value in the gap and the near-wall region of the edge sub-channel which is induced by the axial velocity gradient. Significant intensity of lateral velocity fluctuation is observed which indicates the strong lateral mixing in a 3×3 rod bundle. Through the correlation analysis coherent structures have been observed in the gap region. The spectral analysis shows that the LDV measurement complies to the Komogorov spectrum law, f−5/3, well. The low-frequency peak spectral density of the axial velocity fluctuation has been observed in the gap region connecting sub-channels with velocity difference. The performance of the SSG model and the baseline Reynolds stress model are investigated based on the experiment result. The models predict higher axial velocity in the interior sub-channel and lower in the edge and corner ones than the experiment result. Large discrepancy between the calculated and measured axial flow velocity is resulted from failure in calculating the strong negative u′w′¯ in the gap region connecting different sub-channels.

Influence of grid type and turbulence model on the numerical prediction of the flow around marine propellers working in uniform inflow

In this work we analyze the influence of grid type and turbulence model on the numerical prediction of the flow around marine propellers, working in uniform inflow. The study is carried out comparing hexa-structured meshes with hybrid-unstructured meshes using the SST (Shear Stress Transport) turbulence model and the BSL-RSM (Baseline-Reynolds Stress Model) turbulence model. The simulations are carried out with a commercial CFD solver. The numerical results are compared with the available experimental data of two propellers in model scale. The comparison is carried out evaluating the global field values, represented by the thrust and torque coefficients, and also considering some local field values measured in the propeller wake. The computational results suggest that, for the numerical predictions of the propeller open water propulsion characteristics, the hexa-structured and hybrid-unstructured meshes can guarantee similar levels of accuracy. Nevertheless hybrid-unstructured meshes seem to exhibit a more diffusive character than hexa-structured meshes, and thus they are less suited for detailed investigations of the flow field. Finally, the two different turbulence models behave similarly on both types of meshes, with the BSL-RSM turbulence model providing only slightly better predictions than the computationally more economical SST turbulence model.

CFD analyses of liquid metal flow in sub-channels for Gen IV reactors

Some of the limitations of Reynolds Averaged Navier Stokes (RANS) based Computational Fluid Dynamics CFD codes in computing the flow and temperature field in a rod-bundle are well known. An in-house validation campaign has indicated that the Baseline Reynolds Stress Model (BSL-RSM) with automatic wall treatment is preferred for RANS analyses of a rod-bundle using the CFX code. As a first step in the present paper, the employed CFX code has been assessed with the analyses of a liquid sodium flow in a rod-bundle as in the TEGENA (TEmperatur- und GEschwindigkeitsverteilungen in Stabbündel mit turbulenter NAtriumströmung) experiment. For this RANS analysis, the full cross-section is modelled to avoid numerical issues associated with symmetric boundary conditions. The influence of pitch-to-diameter ratio ( p/ d) and rod arrangements on thermal-hydraulics is analyzed by applying the assessed modeling approach. For this purpose, rod-bundles with different p/ d are arranged in a square and triangular lattice. The computational sub-channels make use of periodic boundary conditions. RANS computed axial velocity normalized with the friction velocity shows the presence of a logarithmic outer region for both arrangements. Similar behavior was reported based on a Large Eddy Simulation (LES) approach. The analyses reveal that the intensity of secondary flow increases with decreasing p/ d for both arrangements. RANS analyzed normal Reynolds stresses normalized with centerline velocity in the smallest gap of rod-bundle reveal their anisotropy. Furthermore, the analyses show that the Nusselt numbers increase with p/d for described flow conditions and for both arrangements. Following observations of flow oscillations in a tight lattice rod-bundle as in Hooper's experiment, as a final step, unsteady RANS simulations for hydraulic analyses using a rod-bundle with small p/ d are presented with two commercial CFD codes, namely, CFX and STAR-CCM+. In particular, the analysis of Hooper's hydraulics experiment with a tight lattice rod-bundle having a p/ d of 1.1 demonstrates the existence of flow oscillations or instabilities as inferred in the experiment.

A stepwise development and validation of a RANS based CFD modelling approach for the hydraulic and thermal-hydraulic analyses of liquid metal flow in a fuel assembly

The demand of reliable and clean energy at affordable prices poses a formidable challenge to the world. The initiatives from various international organizations reserve an important role for liquid metal cooled reactor systems. Assessment of such reactors usually involves unconventional thermal-hydraulics. Consequently, Reynolds Averaged Navier Stokes (RANS) based Computational Fluid Dynamics (CFD) approaches are expected to play a vital role along with various ongoing experiments for the design and the safe operation of these nuclear reactors. One of the major issues is the heat transport in the fuel assembly by the liquid metal. The known difficulties in heat transfer experiments, especially with liquid metals, necessitate the application of RANS in computing details of flow and temperature distribution. Considering these aspects, a four step approach is described in the current paper. As a first step, the heat transfer in a liquid metal flow inside a heated tube is analyzed using a RANS approach and then compared with some of the empirical correlations. The computed Nusselt number reveals the required development length of the thermal boundary layer in liquid metal. Furthermore, these simulations reveal the need of further assessment of this approach and all the existing correlations, and the care that should be taken while applying one of these correlations. In the second step, numerical simulations of the flow of heavy liquid metal around a heated rod in an annular cavity confirm that a RANS strategy can be employed in liquid metal flows. Furthermore, a comparison between computed and experimental non-dimensional axial temperature at the heated rod surface shows that among the considered turbulence models the use of a Baseline-Reynolds Stress Model (BSL-RSM) with automatic wall treatment (AWT) can be preferred for complex geometries. This is also demonstrated in the third step by computing the flow distribution in a triangular arrangement of a fuel assembly and by comparing with an existing hydraulics experiment in rod-bundle. These analyses reveal that the use of the BSL-RSM turbulence model with AWT allows prediction of the cross-flow in this rod-bundle. As the forth and last step, the integral TEGENA (TEmperatur- und GEschwindigkeitsverteilungen in Stabbündel mit turbulenter NAtriumströmung) experiment has been selected for further assessment of RANS based CFD approach in computing both the flow and temperature field. A comparison with literature demonstrates that the use of symmetric boundary condition in such a tightly packed parallel rod-bundle leads to a distorted flow field. The experimentally and the computationally obtained temperature field at a plane in the outlet reveal its acceptable predictive capability. Furthermore, application of two different Reynolds Stress Models yields almost the same temperature distribution as a result of the use of simple first-order gradient model for the turbulent heat fluxes. Consequently, these four steps support the use of this modelling approach for investigating the heat transport in (heavy) liquid metals. Finally, the preferred RANS approach has been applied for thermal-hydraulic evaluation in the square arrangement of a bare rod-bundle as is to be employed in the European Lead-cooled reactor System (ELSY). Also, the influence of rod pitch-to-diameter ratio has been analyzed by numerically re-arranging these rods in the different square lattice. Moreover, arranging the bare rods in triangular lattice at the different rod pitch-to-diameter ratios shows the effect of the square and triangular lattice in thermal-hydraulics. Lastly, comparisons with some of the existing heat transfer correlations for the triangular and the rectangular lattice allows us to identify preferred correlations for these lattice arrangements of bare and liquid metal cooled rod-bundles.

Contribution to single-point closure Reynolds-stress modelling of inhomogeneous flow

This paper is concerned with recent advances in the development of near wall-normal-free Reynolds-stress models, whose single point closure formulation, based on the inhomogeneity direction concept, is completely independent of the distance from the wall, and of the normal to the wall direction. In the present approach the direction of the inhomogeneity unit vector is decoupled from the coefficient functions of the inhomogeneous terms. A study of the relative influence of the particular closures used for the rapid redistribution terms and for the turbulent diffusion is undertaken, through comparison with measurements, and with a baseline Reynolds-stress model (RSM) using geometric wall normals. It is shown that wall-normal-free rsms can be reformulated as a projection on a tensorial basis that includes the inhomogeneity direction unit vector, suggesting that the theory of the redistribution tensor closure should be revised by taking into account inhomogeneity effects in the tensorial integrity basis used for its representation.