Mean Velocity Field Research Articles

Pre-processing DNS data to improve statistical convergence and accuracy of mean velocity fields in invariant data-driven turbulence models

Pre-processing DNS data to improve statistical convergence and accuracy of mean velocity fields in invariant data-driven turbulence models

Influence of aspect ratio on the wake dynamics of a pitching wing

The three-dimensional flow around a custom wing performing pitching oscillation is analyzed numerically using large-eddy simulation with the wall adapting local eddy-viscosity subgrid-scale model. The wake characteristics behind the pitching wing with three different aspect ratios, AR=2/3, 1, 4/3, is investigated in an upstream flow with Reynolds number, Re=2640. The Q−function is employed to identify the swirling structures in the flow field, and four primary vortices, namely, primary leading edge vortex (PLEV), secondary leading edge vortex, wing tip vortices, and trailing edge vortex, are identified. The evolution of vortices over the wing surface is presented in detail, and the arrangement of vortex rings in the wake is compared for three different aspect ratios. The PLEV develops into an arch vortex using the self-induction phenomenon, and the interlinking between the subsequent vortex rings depends on the arch vortex leaving the trailing edge. The force coefficient computed over the wing shows a positive thrust for AR=4/3. The mean velocity field contours indicate a spanwise loading on the wing, which is unaccountable in two-dimensional studies. The Reynolds stresses generated due to large-scale periodic velocity fluctuations in the wake are presented. The vorticity field in the midplane of the wing appears to be diffusive at higher aspect ratios. The comparison of vorticity from the three-dimensional and two-dimensional models suggests that a two-dimensional model will over predict the vorticity distribution and force coefficients in the wake.

Aerodynamic Interactions of Side-by-Side Rotors in Ground Proximity

An experimental investigation is conducted to study the aerodynamic behavior of a two-rotor system in ground proximity. The counter-rotating rotors are placed side-by-side in the hovering condition. The time-averaged and unsteady flow behavior is studied when the rotor-to-rotor lateral distance and the distance between the rotors and the ground are varied. The experiments are performed using three-dimensional large-scale volumetric velocimetry with helium-filled soap bubbles as tracers, tracked by the particle motion analysis technique “Shake-The-Box.” The mean velocity field reveals the wake deflection due to the ground plane and the formation of toroidal-shape regions of separated flow below each rotor. The interaction of the wall jets formed by slipstream deflection results in a separation line with the flow emerging from the wall in a fountain-like pattern. Regimes of flow re-ingestion occur when the rotors are sufficiently far apart. The flowfield exhibits the tendency toward asymmetric states, during which the fountain flow column and the domain of re-ingestion shift closer to one of the rotors. A generic classification of flow regimes is proposed in relation to the behavior of two rotors in ground effect.

Experimental Techniques against RANS Method in a Fully Developed Turbulent Pipe Flow: Evolution of Experimental and Computational Methods for the Study of Turbulence

Fully developed turbulent flow in a pipe was studied by considering experimental and computational methods. The aim of this work was to build on the legacy of the University of Manchester, which is widely regarded as the birthplace of turbulence due to the pioneering work of the prominent academic Professor Osborne Reynolds (1842–1912), by capturing the evolution of fluid turbulence analysis tools over the last 100 years. A classical experimental apparatus was used to measure the mean velocity field and wall shear stress through four historical techniques: static pressure drop; mean square signals measured from a hot-wire; Preston tube; and Clauser plot. Computational Fluid Dynamics (CFD) was used to simulate the pipe flow, utilizing the Reynolds-averaged Navier–Stokes (RANS) method with different two-equation turbulence models. The performance of each approach was assessed to compare the experimental and computational methods. This comparison revealed that the numerical results produced a close agreement with the experiments. The finding shows that, in some cases, CFD simulations could be used as alternative or complementary methods to experimental techniques for analyzing fully developed turbulent pipe flow.

Large-Eddy Simulations of Idealized Shock/Boundary-Layer Interactions with Crossflow

An idealized shock/boundary-layer interaction problem that is two dimensional in the mean but with a three-component mean velocity field is studied using large-eddy simulations. The problem isolates some aspects of three-dimensionality while avoiding others and can thus offer insights into the differences between two- and three-dimensional problems. The addition of a crossflow increases the size of the separation bubble quite substantially when normalized by the boundary-layer thickness or the momentum thickness but less substantially when normalized by the displacement thickness. All cases studied are found to have a very thin precursor separation bubble before the point of true separation that extends for at least one boundary-layer thickness but remains fully within the viscous sublayer. Simulations at different Reynolds numbers show that the flow reattaches after the precursor bubble at sufficiently high Reynolds number.

The Effect of Single-Sided Ribs on Heat Transfer and Pressure Drop Within a Trailing Edge Internal Channel of a Gas Turbine Blade

Abstract Convective cooling in a gas turbine blade internal trailing edge channel is often insufficient at the sharp trailing edge. This study examines convective heat transfer and pressure drop within a simplified nonrotating trailing edge channel. The internal passage has been modeled as a right triangular channel with a 9 deg angle sharp corner. A copper plate was heated from underneath via a uniform heat flux heater and examined via infrared thermography for two cases: smooth wall and single-sided ribbed wall. Nonuniformity in the heat flux due to conduction is corrected by a Reynolds-averaged Navier–Stokes (RANS) conjugate heat transfer calculation, which was validated by the mean velocity, friction factor, and temperature fields from experiments and large eddy simulation (LES). Nusselt number distributions illustrate that surface heat transfer is increased considerably with ribs and coupled with the vortices in the flow. Heat transfer at the sharp corner is increased by more than twofold due to ribs placed at the center of the channel due to secondary flow. The present single-sided ribbed channel utilizes secondary flow toward the corner and is presumed to have better thermal performance than a dual-sided ribbed channel. Thus, it is important to set the appropriate rib length within the channel.

Oxygen enrichment effects on CH4-air turbulent flow characteristics in a coaxial swirl burner

ABSTRACT This paper reports results on the effect of oxygen enrichment on flow and turbulent flames in a coaxial swirl burner. The investigation was performed by S-PIV in CH4-air-O2 non-reacting and reacting flows. This work is the continuation of studies carried out previously by the authors on the same burner, with different new operating conditions. The burner consists of 2 concentric tubes, with a swirler located in the annular part through which the oxidant (air-O2) is supplied. The CH4 gas is delivered into the central tube and injected radially into the combustion chamber through small holes. The burner is placed in a parallelepiped combustion chamber with a volume of 120×50×50 cm3. The paper reports detailed results on mean velocity fields, profiles of the three velocity components (U,V,W) and velocity fluctuations (U’,V’,W’) in the case of a 150 Nl/min oxidant flow rate, a swirl number Sn = 1.4 and an equivalence ratio Φ = 1. Three cases of O2 addition were investigated, 21%, 25%, and 30% (in vol.) corresponding to 9.4 kW, 11.7 kW, and 14.5 kW of combustion power, respectively. In-depth analysis of the field velocity differences between reactive and non-reactive cases was carried out and showed a significant difference in flow velocities between reacting and non-reacting flows. The profiles of the mean velocities, fluctuations and velocity decays differed, and higher values were observed in the reacting case. The enrichment of air with oxygen affected the velocity fields and their fluctuations in both distribution and values. The addition of O2 increased the flame temperature, leading to a greater radial expansion and a reduction in the size of the recirculation zone. The maximum axial velocity increased significantly as a function of the O2 rate, whereas the swirling velocity varied only slightly. The velocity fluctuation results also revealed a difference in profiles and values depending on O2 enrichments and locations along the flow.

Analysis of the three-dimensional swirling and non-swirling jet impingement using a turbulence model with cross-diffusion correction

Three-dimensional swirling impinging jets using the aerodynamic swirl generator are carried out to understand the pure effects of swirl. A shear stress transport model with cross-diffusion correction (SSTCD) is validated for the above flows with the documented experiments. Based on the good performance of the SSTCD model, a detailed flow analysis for Re = 23,000 and swirl numbers of 0 and 0.45 at nozzle-plate spacings of 2 and 6 is presented in terms of the flow structures, mean velocity field, wall shear stress and heat transfer. The results show that the strength of the swirl is reduced with increasing the nozzle-plate spacing near the wall. On the contrary, the effect of swirl on both flow fields and heat transfer is more evident at high nozzle-plate spacing, leading to a broader impact region along the impinging plate. However, downstream, the effect of swirl can be ignored for the above fields at all nozzle-plate spacings. In addition, the swirl motion contributes to high turbulence but the heat transfer rate still decreases with an increase of swirl number. Meanwhile, the uniformity of heat transfer is improved by the swirl motion for both nozzle-plate spacings, except that near the stagnation region at H/D = 2. It is also found that this improvement depends on the Reynolds number.

Numerical analysis of the flow in a cylindrical container injected air from a slit nozzle

This paper describes the flow characteristics of air injected from a slit nozzle in a cylindrical container. The numerical simulation was performed by the large-eddy simulation (LES) using ANSYS Fluent 19.0 based on the finite-volume method. The non-dimensional nozzle offset distance was set to yn/D = 0.4, where D is the diameter of the cylindrical container. The Reynolds number of the plane jet flow was settled at 2,000. The velocity profiles by the numerical simulation agreed with the experimental results. The jet reached near the bottom of container, and the jet flowed out from the left side of container. Although the flow in the cylindrical container showed a three-dimensional turbulent flow structure, the mean velocity field was generated symmetrically with respect to the central cross-section of the container perpendicular to the slit nozzle. The large recirculating flow around the z-axis occurred near the central cross-section of the container by the jet. In addition, the weak recirculated flows around the y-axis were observed near the upper side wall in the container.

Estimation of pollutant dispersion around a building within non-isothermal boundary-layer using detached eddy simulation

The paper presents the numerical results of modelling pollutant transport from a low source behind a bluff-body imitating a building within a non-isothermal boundary-layer. The main goal of the study is to estimate the tracer gas dispersion in a complex turbulent separated flow behind a building in the presence of interference of the atmospheric boundary-layer and local flows. In the first part of the study we compare numerical approaches URANS and IDDES for turbulent flow prediction on a configuration for which experimental data are available. It is shown that detached eddy simulation approach predicts correctly the main separated flow features and demonstrates a reliable correlation with the experimental data on mean velocity, pollutant concentration and temperature fields. In the second part of the study, the influence of unstable thermal stratified flow on the tracer gas transport around a building is analyzed using IDDES method. The unstable thermal flow regime considered in the study affects the distribution of the pollutant concentration in the re-circulation zone behind the building. The presence of additional buoyancy effects leads to an increase in the gas concentration on the leeward wall of the body and gas transport from a ground region to a height greater than in the case with the neutral boundary-layer.

Automatic segmentation of the great arteries for computational hemodynamic assessment

BackgroundComputational fluid dynamics (CFD) is increasingly used for the assessment of blood flow conditions in patients with congenital heart disease (CHD). This requires patient-specific anatomy, typically obtained from segmented 3D cardiovascular magnetic resonance (CMR) images. However, segmentation is time-consuming and requires expert input. This study aims to develop and validate a machine learning (ML) method for segmentation of the aorta and pulmonary arteries for CFD studies.Methods90 CHD patients were retrospectively selected for this study. 3D CMR images were manually segmented to obtain ground-truth (GT) background, aorta and pulmonary artery labels. These were used to train and optimize a U-Net model, using a 70-10-10 train-validation-test split. Segmentation performance was primarily evaluated using Dice score. CFD simulations were set up from GT and ML segmentations using a semi-automatic meshing and simulation pipeline. Mean pressure and velocity fields across 99 planes along the vessel centrelines were extracted, and a mean average percentage error (MAPE) was calculated for each vessel pair (ML vs GT). A second observer (SO) segmented the test dataset for assessment of inter-observer variability. Friedman tests were used to compare ML vs GT, SO vs GT and ML vs SO metrics, and pressure/velocity field errors.ResultsThe network’s Dice score (ML vs GT) was 0.945 (interquartile range: 0.929–0.955) for the aorta and 0.885 (0.851–0.899) for the pulmonary arteries. Differences with the inter-observer Dice score (SO vs GT) and ML vs SO Dice scores were not statistically significant for either aorta or pulmonary arteries (p = 0.741, p = 0.061). The ML vs GT MAPEs for pressure and velocity in the aorta were 10.1% (8.5–15.7%) and 4.1% (3.1–6.9%), respectively, and for the pulmonary arteries 14.6% (11.5–23.2%) and 6.3% (4.3–7.9%), respectively. Inter-observer (SO vs GT) and ML vs SO pressure and velocity MAPEs were of a similar magnitude to ML vs GT (p > 0.2).ConclusionsML can successfully segment the great vessels for CFD, with errors similar to inter-observer variability. This fast, automatic method reduces the time and effort needed for CFD analysis, making it more attractive for routine clinical use.

Drag decomposition of laminar channel flows developing over convergent–divergent riblets

Convergent–divergent (C–D) riblets are known to induce large-scale secondary flow motion in laminar flow that may point to their potential for flow separation control and heat transfer enhancement. In this paper, a fully-developed pressure-driven laminar channel flow with C–D riblets is studied numerically. The effects of Reynolds number, riblet wavelength and riblet cross-sectional shape on the secondary flow and drag characteristics are investigated. An exact expression for the drag coefficient of this flow is derived based on the balance of kinetic energy principle for a rough-wall channel. Furthermore, the triple decomposition technique is used to identify the contribution to drag production from the mean velocity field, the small- and large-scale dispersive flow field. Our results show that the normalized drag increment starts to rise when the secondary flow begins to alter the velocity field across the span as the Reynolds number based on the channel height (ReH) increases above 100, a behavior which is very different from that of a laminar channel flow developing over homogeneous roughness in which the normalized drag increment is found to remain constant up to a much greater Reynolds number due to an absence of secondary flow. It is found that the drag produced by C–D riblets is predominantly caused by the wall-normal gradient of streamwise velocity. In addition, while the drag increment is predominantly caused by the mean and small-scale dispersive velocity at ReH≤100, the contribution from the large-scale dispersive velocity field increases rapidly with the Reynolds number and becomes dominant as ReH exceeds 400. It is also observed that the secondary flow scales with the riblet wavelength Λ, and its swirling strength peaks around Λ=1.25H, which is accompanied by a drag maxima at ReH≥400. Finally, among the three C–D riblet cross-sectional shape (rectangular, triangular and sinusoidal) examined here, the triangular riblet pattern is found to produce a secondary motion with a similar strength to the other two shapes but with less drag penalty.

A Study of Recirculating Flow Fields Downstream of a Diverse Range of Axisymmetric Bluff Body Geometries Suitable for Flame Stabilization

This work investigates the non-reacting time averaged and fluctuating flow field characteristics downstream of a variety of axisymmetric baffles, operating in combination with an upstream double-cavity premixer arrangement. The study aims to broaden knowledge with respect to the impact of different bluff body shapes, leading and trailing edge flow contours, blockage ratios and incoming flow profiles impinging on the bluff body, on the development and properties of the downstream recirculating wake. Particle Image Velocimetry (PIV) measurements have been employed to obtain the mean and turbulent velocity fields throughout the centrally located recirculation zone and the adjacent developing toroidal shear layer. The results are helpful in demarcating the cold flow structure variations in the near wake of the examined baffles which support and, to some extent, determine the flame anchoring performance and heat release disposition in counterpart reacting configurations. Additionally, such results could also assist in the selection of the most suitable flame stabilization configuration for fuels possessing challenging combustion behavior such as multi-component heavier hydrocarbons, biofuels, or hydrogen blends.

Measurement and Numerical Simulation of the Velocity Profile in the Thin Film of an Impinging Water Jet

Abstract Impinging circular free-surface water jets are used in challenging cooling and cleaning tasks. In order to develop simulation models for process optimization, validation data are required, which are currently not available. Therefore, the flow field of these jets is studied for the first time with the novel laser Doppler velocity profile sensor. The mean velocity field and fluctuations are measured within the stagnation and adjacent redirection region for radial coordinates up to three times the nozzle diameter. In the examined parameter range with jet velocities up to 17 m/s and nozzle diameters up to 5.2 mm, i.e., Reynolds numbers up to 69 500, thin films of a few hundred micrometers are formed, which hinder the measurement with common optical measuring systems. Based on the measurement results, a comparatively low-cost volume of fluid simulation model is developed and validated that presumes a relaminarized film flow. The profiles measured and the simulated flow show very good agreement. In the future, the simulation model provides a basis for process optimization and the innovative measurement technology used will prospectively provide further detailed insights into other flows with high velocity gradients.

Error propagation and conditioning analysis of DNS data of turbulent viscoelastic channel flows

In the context of drag reduction phenomenon, direct numerical simulation (DNS) of turbulent viscoelastic flows is a very useful tool able to provide information that is not possible with experiments and that can be used for modeling purposes. A rare and challenging task is to provide measures for statistical errors that are present in DNS data. In the present communication, we employ a method to quantify the statistical convergence of the DNS turbulent viscoelastic channel flows from the unbalance on the mean momentum equation. In addition, we conduct a conditioning analysis of the resultant numerical linear system of this problem. We compare results from different research groups. We found that DNS databases of viscoelastic flows can induce high error propagation in the mean velocity. As a consequence, larger sampling times are needed to achieve a statistically converged Reynolds stress tensor. The conditioning analysis has shown that elasticity enhances conditioning. • We discuss the DNS importance for turbulence modeling. • Uncertainty and statistical error are analyzed in the DNS context. • Equations for mean viscoelastic turbulence (VT) are ill-conditioned. • Error propagation in the mean velocity field related to VT is high. • Statistical quantities, like Reynolds stress, in VT need to be more converged.

Experimental investigation of the wake characteristics behind twin vertical axis turbines

Vertical axis wind and tidal turbines are a promising technology, well suited to harness kinetic energy from highly turbulent environments such as urban areas or rivers. The power density per occupied land area of two or three vertical axis rotors deployed in close proximity can notably exceed that of their horizontal axis counterparts. Using acoustic Doppler velocimetry, the three-dimensional wake developed downstream of standalone and twin vertical axis turbines of various shaft-to-shaft distances and rotational direction combinations was characterised in terms of mean velocity and turbulence statistics, with their impact on momentum recovery quantified. Results show that the wake hydrodynamics were more impacted by turbine rotational direction than lateral distance between devices for the range of lateral spacing considered. In the cases with turbines operating in a counter-rotating forward configuration, the wake mostly expanded laterally and attained the largest velocities that exceeded those in the single turbine case, with full momentum recovery at 5 turbine diameters downstream. The wake developed by the counter-rotating backward setup notably extended over the vertical direction, whilst devices rotating in the same direction featured the greatest lateral wake expansion with reduced velocities. Linear wake superposition of the single turbine wake provided a good representation of the mean velocity field behind twin-turbine setups. The presented results indicate that, in the design of twin-turbine arrays moving in counter-rotating forward direction, a lateral spacing of, at least, two turbine diameters should be kept as this allows the kinetic energy in the wake to be fully recovered by five turbine diameters downstream.

The effects of swirl vanes and a vortex stabilizer on the dynamic flow field in a cyclonic separator

The present work describes the isolated and combined effects of two internal components – swirl vanes and a vortex stabilizer – on the properties of the flow inside a particular type of gas–liquid cyclonic separator. These elements play an important role in the optimization of the catch efficiency of the gas and liquid phases but may otherwise result in a large increase in pressure drop. It is shown that the vanes and the vortex stabilizer have a considerable influence on the position of the center of rotation of the flow, which clearly does not coincide with the geometric center of the cyclone. Here, an experimental study investigates the velocity field and the resulting pressure drop inside the cyclone to account for the individual effects caused by the two internal components. Particle Image Velocimetry is used to characterize the mean velocity field and the precession of the vortex core (PVC). For the different geometrical configurations discussed in the present work, the influence of the swirl vanes was always beneficial to reduce the residence time and thus decrease the pressure drop across the cyclonic separator. The present work also introduces a method for the assessment of the PVC frequency of a confined swirling flow for low frequency 2D-PIV measurements.

Turbulence Measurements at the Inlet of an Automotive Turbocharger Compressor Using Stereoscopic Particle Image Velocimetry

Abstract The present work uses stereoscopic particle image velocimetry (SPIV) to analyze the compressor inlet flow field, with specific emphasis on its turbulence characteristics during flow reversal in order to gain further insight into the inlet flow structures. SPIV experiments were carried out at the inlet of a centrifugal compressor without any recirculation channel at four different rotational speeds (from 80 to 140 krpm) and over the entire mass flow range (from choke to surge) at each speed. Detailed analyses have been carried out for the mean velocity field, the mean vorticity field, and the turbulent statistics including turbulent kinetic energy, Reynolds stress, and the one-dimensional energy spectra. The turbulent kinetic energy at the compressor inlet was observed to increase rapidly along a speed line with decreasing mass flowrate once flow reversal started, and the turbulence became more anisotropic. As the flowrate was reduced (along a speed line), the zone with maximum turbulent kinetic energy moved from the periphery toward the center of the inlet duct and also occurred further upstream from the impeller. The Reynolds stress distributions suggest that the Boussinesq assumption of an isotropic eddy viscosity may not be appropriate after the detection of flow reversal. The Reynolds shear stresses were observed to change signs with their corresponding velocity gradients at the tested mass flowrates at different rotational speeds. At the investigated flowrates, the radial gradients in the axial and tangential velocities were found to be most dominant.

Flow Field Parametric Interpolation Using a Proper Orthogonal Decomposition: Application to the Variable Valve Timing Effect on a Tumble In-cylinder Miller Engine Mean Flow

The current article presents a method to reconstruct the mean velocity field of a cyclic flow for an input parameter value that has not been measured, allowing for the number of tests to be reduced. It is applied to the tumble flow of a gasoline engine following a Miller cycle. New engines often include variable valve timing (VVT) systems to maximize the efficiency of such over-expanded cycles for different operating points. The reconstruction was thus carried out using different offset values of the intake valve lift timing. Experimental data were collected from a transparent engine in an early intake valve closing (EIVC) configuration using particle image velocimetry (PIV). The mean velocity field reconstruction was based on the interpolation of the proper orthogonal decomposition (POD) coefficients. The accuracy of the method was evaluated at different points by comparing the interpolated and the measured flow fields. The accuracy was estimated by calculating the error in the rotation rate of the tumble and the position of its center of rotation. The new mean velocity field set allowed for the position of the tumble’s center of rotation to be closely tracked according to the input parameter and a rotation rate map to be made. Some results on Miller’s cycle could thus be found and the data generated could guide future developments.

Characterisation of vortex shedding on a hydrofoil using PIV measurements

An experimental study of vortex shedding on a hydrofoil Eppler 817 was conducted using two-dimensional two components Particle Image Velocimetry. This foil section’s characteristics are adapted for naval applications but sparsely documented. The characterization of the flow modes was realized based on statistical data such as the mean velocity field and the standard deviation of the vertical velocities. The data were acquired at very low Reynolds number which are not often covered for such hydrofoil and at four angles of attack ranging from 2◦ to 30◦. A map of different characteristic flow modes was made for this space of parameters and was used to identify flow configurations exhibiting particular dynamics.