Hybrid RANS/LES computations of plane impinging jets with DES and PANS models

The qualities of a DES (Detached Eddy Simulation) and a PANS (Partially-Averaged Navier–Stokes) hybrid RANS/LES model, both based on the k–ω RANS turbulence model of Wilcox (2008, “Formulation of the k–ω turbulence model revisited” AIAA J., 46: 2823–2838), are analysed for simulation of plane impinging jets at a high nozzle-plate distance (H/B = 10, Re = 13,500; H is nozzle-plate distance, B is slot width; Reynolds number based on slot width and maximum velocity at nozzle exit) and a low nozzle-plate distance (H/B = 4, Re = 20,000). The mean velocity field, fluctuating velocity components, Reynolds stresses and skin friction at the impingement plate are compared with experimental data and LES (Large Eddy Simulation) results. The k–ω DES model is a double substitution type, following Davidson and Peng (2003, “Hybrid LES–RANS modelling: a one-equation SGS model combined with a k–ω model for predicting recirculating flows” Int. J. Numer. Meth. Fluids, 43: 1003–1018). This means that the turbulent length scale is replaced by the grid size in the destruction term of the k-equation and in the eddy viscosity formula. The k–ω PANS model is derived following Girimaji (2006, “Partially-Averaged Navier–Stokes model for turbulence: a Reynolds-Averaged Navier–Stokes to Direct Numerical Simulation bridging method” J. Appl. Mech., 73: 413–421). The turbulent length scale in the PANS model is constructed from the total turbulent kinetic energy and the sub-filter dissipation rate. Both hybrid models change between RANS (Reynolds-Averaged Navier–Stokes) and LES based on the cube root of the cell volume. The hybrid techniques, in contrast to RANS, are able to reproduce the turbulent flow dynamics in the shear layers of the impacting jet. The change from RANS to LES is much slower however for the PANS model than for the DES model on fine enough grids. This delays the break-up process of the vortices generated in the shear layers with as a consequence that the DES model produces better results than the PANS model.     

Numerical study of oscillating boundary layer flow over a flat plate using k–kL–ω turbulence model

Oscillating boundary layer flow over an infinite flat plate at rest was simulated using the k–k_L–ω turbulence model for a Reynolds number range of 32 ⩽ Re_δ ⩽ 10,000 ranging from fully laminar flow to fully turbulent flow. The k–k_L–ω model was validated by comparing the predictions with LES results and experimental results for intermittently turbulent and fully turbulent flow regimes. The good agreement obtained between the k–k_L–ω model prediction with the experimental and LES results indicate that the k–k_L–ω model is able to accurately simulate transient intermittently turbulent flow and as well as accurately predict the onset of turbulence for such oscillatory flows.     

Flow structures and heat transfer on dimples in a staggered arrangement

Vortex structures and heat transfer enhancement mechanism of turbulent flow over a staggered array of dimples in a narrow channel have been investigated using Large Eddy Simulation (LES), Laser Doppler Velocimetry (LDV) and pressure measurements for Reynolds numbers Re_H = 6521 and Re_H = 13,042.The flow and temperature fields are calculated by LES using dynamic mixed model applied both for the velocity and temperature. Simulations have been validated with experimental data obtained for smooth and dimpled channels and empiric correlations. The flow structures determined by LES inside the dimple are chaotic and consist of small eddies with a broad range of scales where coherent structures are hardly to detect. Proper Orthogonal Decomposition (POD) method is applied on resolved LES fields of pressure and velocity to identify spatial–temporal structures hidden in the random fluctuations. For both Reynolds numbers it was found that the dimple package with a depth h to diameter D ratio of h/D = 0.26 provides the maximum thermo-hydraulic performance. The heat transfer rate could be enhanced up to 201% compared to a smooth channel.     

Reconstruction of turbulent fluctuations for hybrid RANS/LES simulations using a Synthetic-Eddy Method

A coupling methodology between an upstream Reynolds Averaged Navier–Stokes (RANS) simulation and a Large Eddy Simulation (LES) further downstream is presented. The focus of this work is on the RANS-to-LES interface inside an attached turbulent boundary layer, where an unsteady LES content has to be explicitly generated from a steady RANS solution. The performance of the Synthetic-Eddy Method (SEM), which generates realistic synthetic eddies at the inflow of the LES, is investigated on a wide variety of turbulent flows, from simple channel and square duct flows to the flow over an airfoil trailing edge. The SEM is compared to other existing methods of generation of synthetic turbulence for LES, and is shown to reduce substantially the distance required to develop realistic turbulence downstream of the inlet.     

DNS and LES of turbulent flow in a closed channel featuring a pattern of hemispherical roughness elements

Direct Numerical Simulations (DNS) and Large Eddy Simulations (LES) were performed for fully-developed turbulent flow in channels with smooth walls and walls featuring hemispherical roughness elements at shear Reynolds numbers Re_τ = 180 and 400, with the goal of studying the effect of these roughness elements on the wall-layer structure and on the friction factor. The LES and DNS approaches were verified first by comparison with existing DNS databases for smooth walls. Then, a parametric study for the hemispherical roughness elements was conducted, including the effects of shear Reynolds number, normalized roughness height (k⁺ = 10–20) and relative roughness spacing (s⁺/k⁺ = 2–6). The sensitivity study also included the effect of distribution pattern (regular square lattice vs. random pattern) of the roughness elements on the walls. The hemispherical roughness elements generate turbulence, thus increasing the friction factor with respect to the smooth-wall case, and causing a downward shift in the mean velocity profiles. The simulations revealed that the friction factor decreases with increasing Reynolds number and roughness spacing, and increases strongly with increasing roughness height. The effect of random element distribution on friction factor and mean velocities is however weak. In all cases, there is a clear cut between the inner layer near the wall, which is affected by the presence of the roughness elements, and the outer layer, which remains relatively unaffected. The study reveals that the presence of roughness elements of this shape promotes locally the instantaneous flow motion in the lateral direction in the wall layer, causing a transfer of energy from the streamwise Reynolds stress to the lateral component. The study indicates also that the coherent structures developing in the wall layer are rather similar to the smooth case but are lifted up by almost a constant wall-unit shift y⁺ (∼10–15), which, interestingly, corresponds to the relative roughness k⁺ = 10.     

Turbulent Couette–Taylor flows with endwall effects: A numerical benchmark

The accurate prediction of fluid flow within rotating systems has a primary role for the reliability and performance of rotating machineries. The selection of a suitable model to account for the effects of turbulence on such complex flows remains an open issue in the literature. This paper reports a numerical benchmark of different approaches available within commercial CFD solvers together with results obtained by means of in-house developed or open-source available research codes exploiting a suitable Reynolds Stress Model (RSM) closure, Large Eddy Simulation (LES) and a direct numerical simulation (DNS). The predictions are compared to the experimental data of Burin et al. (2010) in an original enclosed Couette–Taylor apparatus with endcap rings. The results are discussed in details for both the mean and turbulent fields. A particular attention has been turned to the scaling of the turbulent angular momentum G with the Reynolds number Re. By DNS, G is found to be proportional to Re^α, the exponent α = 1.9 being constant in our case for the whole range of Reynolds numbers. Most of the approaches predict quite well the good trends apart from the k–ω SST model, which provides relatively poor agreement with the experiments even for the mean tangential velocity profile. Among the RANS models, even though no approach appears to be fully satisfactory, the RSM closure offers the best overall agreement.     

Extending the bounds of ‘steady’ RANS closures: Toward an instability-sensitive Reynolds stress model

The incapability of the conventional Unsteady RANS (Reynolds–Averaged Navier Stokes) models to adequately capture turbulence unsteadiness presents the prime motivation of the present work, which focuses on formulating an instability-sensitive, eddy-resolving turbulence model on the Second-Moment Closure level. The model scheme adopted, functioning as a ‘sub-scale’ model in the Unsteady RANS framework, represents a differential near-wall Reynolds stress model formulated in conjunction with the scale-supplying equation governing the homogeneous part of the inverse turbulent time scale ω_h (ω_h = ɛ_h/k). The latter equation was straightforwardly obtained from the model equation describing the dynamics of the homogeneous part of the total viscous dissipation rate ɛ, defined as ɛ_h = ɛ − 0.5ν∂²k/(∂x_j∂x_j) (Jakirlic and Hanjalic, 2002), by applying the derivation rules to the expression for ω_h. The model capability to account for vortex length and time scales variability was enabled through an additional term in the corresponding length-scale determining equation, providing a selective enhancement of its production, pertinent particularly to the highly unsteady separated shear layer region, modeled in terms of the von Karman length scale (comprising the second derivative of the velocity field) in line with the SAS (Scale-Adaptive Simulation) proposal (Menter and Egorov, 2010). The present model formulation, termed as SRANS model (Sensitized RANS), does not comprise any parameter depending explicitly on grid spacing. The predictive capabilities of the newly proposed length-scale determining model equation, solved in conjunction with Jakirlic and Hanjalic’s (2002) Reynolds stress model equation, are presently demonstrated by computing the flow configurations of increasing complexity featured by boundary layer separation from sharp-edged and continuous curved surfaces: backward-facing step flow, flow over a wall-mounted fence, flow over smoothly contoured periodically arranged hills and flow in a 3-D diffuser. The model performances are also assessed in capturing the natural decay of the homogeneous isotropic turbulence and the near-wall Reynolds stress anisotropy in a plane channel. In most cases considered the fluctuating velocity field was obtained starting from steady RANS results.     

LES of the flow around several cuboids in a row

This paper presents Large Eddy Simulations (LES) of flow around a four-vehicle platoon when one of the platoon members was forced to undergo in-line oscillations. The LES were made at the Reynolds number of 10⁵ based on the height of the vehicles. Combinations of two different frequencies corresponding to non-dimensional frequencies at the Strouhal numbers St₁ = 0.025 and St₂ = 0.013 and two oscillation amplitudes were used in this study. The methodology was validated by comparisons with data from previous experimental investigations. In order to highlight the dynamic effects, comparisons were made with steady results on a single vehicle and on a four-vehicle platoon. Large differences were found in the flow structures between quasi-steady and dynamic results. Furthermore, the behavior of the drag coefficient of the upstream neighbor of the oscillating model was investigated.     

Large Eddy Simulations of turbulent particle-laden channel flow

This paper scrutinises the Large Eddy Simulation (LES) approach to simulate the behaviour of inter-acting particles in a turbulent channel flow. A series of simulations that are fully (four-way), two-way and one-way coupled are performed in order to investigate the importance of the individual physical phenomena occurring in particle-laden flows. Moreover, the soft sphere and hard sphere models, which describe the interaction between colliding particles, are compared with each other and the drawbacks and advantages of each algorithm are discussed. Different models to describe the sub-grid scale stresses with LES are compared. Finally, simulations accounting for the rough walls of the channel are compared to simulations with smooth walls. The results of the simulations are discussed with the aid of the experimental data of Kussin J. and Sommerfeld M., 2002, Experimental studies on particle behaviour and turbulence modification in horizontal channel flow with different wall roughness, Exp. in Fluids, 33, pp. 143–159 of Reynolds number 42,000 based on the full channel height. The simulations are carried out in a three-dimensional domain of 0.175 m × 0.035 m  × 0.035 m where the direction of gravity is perpendicular to the flow. The simulation results demonstrate that rough walls and inter-particle collisions have an important effect in redistributing the particles across the channel, even for very dilute flows. A new roughness model is proposed which takes into account the fact that a collision in the soft sphere model is fully resolved and it is shown that the new model is in very good agreement with the available experimental data.     

URANS of flow and endwall heat transfer in a pinned passage relevant to gas-turbine blade cooling

This paper presents some results of URANS study of flow and heat transfer in a matrix of wall-bounded 8 × 8 round pins, mimicking internal cooling passage of gas-turbine blades. The focus is on flow unsteadiness, its role in heat transfer and the capabilities of RANS models to reproduce these features in a set-up of industrial relevance. The results for two Reynolds numbers, 10 000 and 30 000, are compared with the available experiments and LES. It is shown that the elliptic-relaxation eddy-viscosity model, ζ-f captures vortex shedding and the consequent gross effects on the flow development. However, a closer look at flow details reveals discrepancies, especially around the first three pin rows, where the unsteadiness reproduced by URANS shows much weaker amplitudes as compared with LES. Only further downstream the succession of forcing from a series of pins produced unsteadiness akin to those captured by LES. The comparison suggests that smaller structures undetected by URANS need to be resolved to capture properly the separation and wake characteristics of each row. At Re = 10 000, the average endwall Nusselt number agrees well with the LES, both being about 20% lower than in the experiment. For Re = 30 000 the URANS Nusselt is within 10% of the experimental value.     

Direct numerical simulation of a NACA0012 in full stall

This work aims at investigating the mechanisms of separation and the transition to turbulence in the separated shear-layer of aerodynamic profiles, while at the same time to gain insight into coherent structures formed in the separated zone at low-to-moderate Reynolds numbers. To do this, direct numerical simulations of the flow past a NACA0012 airfoil at Reynolds numbers Re = 50,000 (based on the free-stream velocity and the airfoil chord) and angles of attack AOA = 9.25° and AOA = 12° have been carried out. At low-to-moderate Reynolds numbers, NACA0012 exhibits a combination of leading-edge/trailing-edge stall which causes the massive separation of the flow on the suction side of the airfoil. The initially laminar shear layer undergoes transition to turbulence and vortices formed are shed forming a von Kármán like vortex street in the airfoil wake. The main characteristics of this flow together with its main features, including power spectra of a set of selected monitoring probes at different positions on the suction side and in the wake of the airfoil are provided and discussed in detail.     

Adverse pressure gradient turbulent flows over rough walls

An experimental study of a fully developed turbulent channel flow and an adverse pressure gradient (APG) turbulent channel flow over smooth and rough walls has been performed using a particle image velocimetry (PIV) technique. The rough walls comprised two-dimensional square ribs of nominal height, k = 3 mm and pitch, p = 2k, 4k and 8k. It was observed that rib roughness enhanced the drag characteristics, and the degree of enhancement increased with increasing pitch. Similarly, rib roughness significantly increased the level of turbulence production, Reynolds stresses and wall-normal transport of turbulence kinetic energy and Reynolds shear stress well beyond the roughness sublayer. On the contrary, the distributions of the eddy viscosity, mixing length and streamwise transport of turbulence kinetic energy and Reynolds shear stress were reduced by wall roughness, especially in the outer layer. Adverse pressure gradient produced a further reduction in the mean velocity (in comparison to the results obtained in the parallel section) but increased the wall-normal extent across which the mean flow above the ribs is spatially inhomogeneous in the streamwise direction. APG also reinforced wall roughness in augmenting the equivalent sand grain roughness height. The combination of wall roughness and APG significantly increased turbulence production and Reynolds stresses except in the immediate vicinity of the rough walls. The transport velocities of the turbulence kinetic energy and Reynolds shear stress were also augmented by APG across most part of the rough-wall boundary layer. Further, APG enhanced the distributions of the eddy viscosity across most of the boundary layer but reduced the mixing length outside the roughness sublayer.     

Particle image velocimetry investigation on mixing enhancement of non-circular sharp edge nozzles

An experimental study using Particle Image Velocimetry (PIV) on free jets issuing from different orifice plate (OP) nozzles is reported. Mean velocity, turbulence intensity and higher order profiles relevant for large and small scale mixing are considered in the near field and interaction zone (0 < X/D < 20). This is done to determine mixing enhancement due to rectangular, squared, elliptic and triangular nozzles in comparison to circular nozzle results in two orthogonal planes. The effect of Reynolds number on the differences among the nozzle shapes is also considered by performing measurements just after laminar–turbulent transition (Re = 8000) and in the fully turbulent regime (Re = 35,000). The results at low Reynolds number show two classes of jets, i.e. at one side, those closer to axial-symmetric conditions, as circular, square and triangular jets, whereas on the other side those with elongated nozzles as rectangular and elliptic. The reason for the different behavior of the latter is connected to the phenomenon of axis-switching which allows a rearrangement of turbulence over the different velocity components and directions. However, for the highest Reynolds number investigated, all nozzles show similar behavior especially in the jet far field (X/D > 10), thus suggesting a significant Reynolds number dependence of the results.     

Separated flow structures around a cylindrical obstacle in a narrow channel

A detailed experimental study is performed on the separated flow structures around a low aspect-ratio circular cylinder (pin-fin) in a practical configuration of liquid cooling channel. Distinctive features of the present arrangement are the confinement of the cylinder at both ends, water flow at low Reynolds numbers (Re = 800, 1800, 2800), very high core flow turbulence and undeveloped boundary layers at the position of the obstacle. The horseshoe vortex system at the junctions between the cylinder and the confining walls and the near wake region behind the obstacle are deeply investigated by means of Particle Image Velocimetry (PIV). Upstream of the cylinder, the horseshoe vortex system turns out to be perturbed by vorticity bursts from the incoming boundary layers, leading to aperiodical vortex oscillations at Re = 800 or to break-away and secondary vorticity eruptions at the higher Reynolds numbers. The flow structures in the near wake show a complex three-dimensional behaviour associated with a peculiar mechanism of spanwise mass transport. High levels of free-stream turbulence trigger an early instabilization of the shear layers and strong Bloor–Gerrard vortices are observed even at Re = 800. Coalescence of these vortices and intense spanwise flow inhibit the alternate primary vortex shedding for time periods whose length and frequency increase as the Reynolds number is reduced. The inhibition of alternate vortex shedding for long time periods is finally related to the very large wake characteristic lengths and to the low velocity fluctuations observed especially at the lowest Reynolds number.     

Effects of rotation on flow in an asymmetric rib-roughened duct: LES study

We report on large-eddy simulations (LES) of fully-developed asymmetric flow in a duct of a rectangular cross-section in which square-sectioned, equally-spaced ribs oriented perpendicular to the flow direction, were mounted on one of the walls. The configuration mimics a passage of internal cooling of a gas-turbine blade. The duct flow at a Reynolds number Re = 15,000 (based on hydraulic diameter D_h and bulk flow velocity U₀) was subjected to clock-wise (stabilising) and anti-clock-wise (destabilising) orthogonal rotation at a moderate rotational number Ro = ΩD_h/U₀ = 0.3, where Ω is the angular velocity. The LES results reproduced well the available experimental results of Coletti et al. (2011) (in the mid-plane adjacent to the ribbed wall) and provided insight into the whole duct complementing the reference PIV measurement. We analyzed the effects of stabilising and destabilising rotation on the flow, vortical structures and turbulence statistics by comparison with the non-rotating case. The analysis includes the identification of depth of penetration of the rib-effects into the bulk flow, influence of flow three-dimensionality and the role of secondary motions, all shown to be strongly affected by the rotation and its direction.     

Nusselt number and friction factor in thermally stratified turbulent channel flow under Non-Oberbeck–Boussinesq conditions

In stably stratified turbulence, computations under Oberbeck–Boussinesq (OB) hypothesis of temperature-independent fluid properties may lead to inaccurate representation of the flow field and to wrong estimates of momentum/heat transfer coefficients. This is clearly assessed here comparing direct numerical simulations of stratified turbulence under OB conditions to simulations under NOB (Non-Oberbeck–Boussinesq) conditions of temperature-dependent fluid viscosity and thermal expansion coefficient. Compared to the OB case, NOB conditions may induce local flow relaminarization with significant variations (up to 30%) of heat and momentum transfer coefficients. Together with DNS results, we propose a phenomenological model (based on turbulent bursts) for heat transfer prediction in stratified turbulence under OB and NOB conditions. Implications of NOB assumptions on mixing efficiency (i.e. flux Richardson number Ri_f) and turbulent Prandtl number (Pr_t) are also discussed. These results are of specific importance in RANS modelling, where the condition Pr_t = 1 is usually assumed (Reynolds analogy). Although this assumption is valid in some situations (i.e. boundary layer, pipe flow) there is uncertainty about its validity for stably-stratified turbulence. We demonstrate that this assumption is inaccurate when NOB effects become significant.     

Superiority of PANS compared to LES in predicting a rudimentary landing gear flow with affordable meshes

Results of simulations of the flow around a rudimentary landing gear are presented in the paper. A newly proposed improved Partially-Averaged Navier–Stokes (PANS) method using k − ε − ζ − f turbulence model is used for prediction of the flow. The results are compared with the experimental data but also with the results of two LES simulations performed using the PANS computational grids. PANS simulations predicted the flow in good agreement with the experimental data. LES predicted a non-physical creation of separation over the front wheels that does not exist in the PANS prediction and was not observed in the experimental oil film. PANS simulations showed low sensitivity to the grid refinement. They show clear advantage compared with the LES simulations when the computational grid is inadequate for resolution of the near-wall flow structures.     

A reformulated synthetic turbulence generation method for a zonal RANS–LES method and its application to zero-pressure gradient boundary layers

A synthetic turbulence generation (STG) method for subsonic and supersonic flows at low and moderate Reynolds numbers to provide inflow distributions of zonal Reynolds-averaged Navier–Stokes (RANS) – large-eddy simulation (LES) methods is presented. The STG method splits the LES inflow region into three planes where a local velocity signal is decomposed from the turbulent flow properties of the upstream RANS solution. Based on the wall-normal position and the local flow Reynolds number, specific length and velocity scales with different vorticity content are imposed at the inlet plane of the boundary layer. The quality of the STG method for incompressible and compressible zero-pressure gradient boundary layers is shown by comparing the zonal RANS–LES data with pure LES, pure RANS, and direct numerical simulation (DNS) solutions. The distributions of the time and spanwise wall-shear stress, Reynolds stress distributions, and two point correlations of the zonal RANS–LES simulations are smooth in the transition region and in good agreement with the pure LES and reference DNS findings. The STG approach reduces the RANS-to-LES transition length to less than four boundary-layer thicknesses.     

LES of turbulent jet in cross-flow: Part 1 – A numerical validation study

The paper presents results of a LES based numerical simulation of the turbulent jet-in-cross-flow (JICF) flowfield, with Reynolds number based on cross-flow velocity and jet diameter Re = 2400 and jet-to-cross-flow velocity ratio of R = 3.3. The JICF flow case has been investigated in great detail, involving conduction of two independent precursor simulations, prior to the main JICF simulation, as the considered case has turbulent inflow conditions on both jet and cross-stream side. The LES results are directly compared to pointwise Laser Doppler Anemometry (LDA) measurements, showing a very good agreement on the level of various statistical quantities in all flow regions but the immediate jet-to-cross-flow exhaustion zone. Several LES computations involving grids of up to 15 million grid points have been conducted, showing no improvement in the agreement between numerical results and measurements, possibly indicating a LDA measurement problem in this particular region.