Mean Pressure Distribution Research Articles

Transient aerodynamics of a square cylinder under downburst-like accelerating flows reproduced in a multiple-fan wind tunnel

Aerodynamic and pressure coefficients of a sectional model of a sharp-edged square cylinder, at zero incidence angle, are experimentally evaluated in accelerating flow conditions through the use of a multiple-fan wind tunnel. Different unsteady flows are reproduced, characterized by flow parameters which dictate the initial and target velocities, as well as the flow acceleration and deceleration, and the temporal spacing between the ramp-up and the ramp-down. The accelerating turbulent flows reproduced during the campaign are consistent with those typical of full-scale downbursts.The ensemble mean of the mean drag coefficient in unsteady conditions reveals to be either comparable or definitely lower than the one relevant to the steady case. The drop seems to increase for higher levels of the acceleration and to be enhanced for the ramp-up conditions, while the ramp-down case suggests a lower level of reduction. This non-symmetric pattern appears to be mitigated when the starting wind velocity increases, reflecting a higher initial Reynolds number. Same comments, but with greater discrepancies compared to the steady conditions, might be addressed for the standard deviation of the lift coefficient connected with vortex shedding. Both these findings point out the presence of fluid-memory effects associated with the acceleration, which might affect the development of vortex shedding and the consequent bluff-body aerodynamics in transient conditions. This is also confirmed by the analyses of the mean pressure distributions. In particular, the detailed scrutiny of specific pressure coefficients reflects their different level of interference with the wake and the model edge.

Computational wind analysis of an open air-inflated membrane structure

AbstractThe current research aimed to obtain mean pressure distribution over an air-inflated membrane structure using Computational Wind Engineering tools. The steady-state analysis applied the Reynolds-Averaged Navier-Stokes equations with the standard turbulence model. The pressure coefficients were compared with former experimental results to validate the numerical solution. Significant errors were detected close to the critical flow separation points when comparing the numerical results with the wind tunnel tests. However, these errors are local, and the numerical methodology provides accurate results in those areas with minor turbulence motion influence. In general, the numerical solution provided good approximation of the pressure coefficient fields.

Experimental Verification of the CFD Model of the Squeeze Film Lifting Effect

The presented study shows the results of the research into the squeeze film levitation phenomena. The system introduced in the investigation is composed of a vibrating surface, air squeeze film, and the surface of the body freely suspended over the film. The use of the CFD (Computational Fluid Dynamics) model used in the system allows us to determine the steady state, periodic behavior of the air film (described by Navier–Stokes, continuity equations, and ideal gas law), and the lifted object dynamics. The model allows us to determine multiple factors, among others, mean film thickness and pressure distribution inside the fluid film. The influence of factors, such as vibration amplitude, frequency, and load on the lifting conditions, was presented. A series of calculations show the levitations height in the range of 5.61 up to 58.12 microns, obtained for masses of samples between 5–20 g, vibration frequency of 5–25 kHz, and the motions amplitude of 0.5–1.5 µm. A series of CFD multivariable calculations for a standing wave inducer were not previously published. The CFD model was validated with the use of experiments on a specially developed test rig. The authors experimentally obtained the height of levitation up to 200 microns.

A multi-fidelity wind surface pressure assessment via machine learning: A high-rise building case

Computational fluid dynamics (CFD) represents an attractive tool for estimating wind pressures and wind loads on high-rise buildings. The CFD analyses can be conducted either by low-fidelity simulations (RANS) or by high-fidelity ones (LES). The low-fidelity model can efficiently estimate wind pressures over a large range of wind directions, but it generally lacks accuracy. On the other hand, the high-fidelity model generally exhibits satisfactory accuracy, yet, the high computational cost can limit the number of approaching wind angles that can be considered. In order to take advantage of the main benefits of these two CFD approaches, a multi-fidelity machine learning framework is investigated that aims to ensure the simulation accuracy while maintaining the computational efficiency. The study shows that the accurate prediction of distributions of mean and rms pressure over a high-rise building for the entire wind rose can be obtained by utilizing only 3 LES-related wind directions. The artificial neural network is shown to perform best among considered machine learning models. Moreover, hyperparameter optimization significantly improves the model predictions, increasing the R2 value in the case of rms pressure by 60%. Dominant and ineffective features are determined that provide a route to solve a similar application more effectively.

On the nonlinear behaviour of oscillatory flow in a high pressure amplitude standing-wave thermoacoustic heat engine

This article deals with the numerical study of the nonlinear features of oscillatory flow in a high pressure amplitude standing wave thermoacoustic heat engine. The overall performance of thermoacoustic devices can deteriorate due to nonlinearities caused by their operation at high pressure amplitudes. Acoustic mass streaming is one of these nonlinearities that can influence the performance of thermoacoustic systems by an undesirable convective heat transfer. Therefore, understanding the nonlinear phenomena within thermoacoustic devices is of fundamental and practical importance. The Navier‐Stokes and energy equations are solved on the structured meshes by resolving thermal and viscous boundary layers utilizing the open-source modelling software OpenFOAM. In particular, the paper looks in more detail at the causes of a non-uniform distribution of mean pressure along the resonator reported previously by both experimental and numerical studies. This is explained by an analogy to a “distributed fluidic pump” manifesting itself through the pumping effects of streaming patterns. Also, the non-linear regime of the mean dynamic pressure shows interesting start-up instabilities which can be influential on achievable efficiencies.

The mean pressure distribution on the free end of a square prism

In this experimental study, the mean pressure distribution was measured on the free end of a surface-mounted finite-height square prism for different aspect ratios, incidence angles, and boundary layer thicknesses. The Reynolds number based on the width of the prism was kept constant at Re = 6.5 × 104, the aspect ratio of the prism was changed from AR = 1 to 11, and the incidence angle was varied in small increments from α = 0° to 45°. The thickness of the boundary layer on the ground plane relative to the prism width was either δ/D = 0.8 or δ/D = 2.6, representing a thin or thick boundary layer. When the prism was oriented at α = 0°, distinct mean pressure distributions were observed for prisms nearly or completely immersed in the boundary layer, and also for the very slender high-aspect ratio prisms. The mean normal force coefficient steadily increased with the aspect ratio but its value decreased in the presence of the thicker boundary layer. When the prism was oriented at α = 45°, strong pressure gradients were encountered near the leading edges for prisms nearly or completely immersed in the boundary layer, and the conical vortex angles increased with aspect ratio. For intermediate aspect ratios, these pressure gradients weakened and the vortex angles became insensitive to AR. When the incidence angle was between 0° and 45°, the asymmetric mean pressure distribution meant that the conical vortices were located at different angles. The position of the conical vortex located more centrally above the free end was more sensitive to changes in aspect ratio. The influence of the boundary layer on the vortex angles was more pronounced for smaller aspect ratios. In addition, the behaviour of the normal force coefficient with incidence became more complex for higher aspect ratios.

Reynolds number effects and outer similarity of pressure fluctuations in turbulent pipe flow

We study the structure of pressure fluctuations in turbulent pipe flow, up to friction Reynolds number Reτ=6000, using standard spectral decomposition and proper orthogonal decomposition (POD). The mean pressure distribution is found to be qualitatively different from the case of channel flow, with an additional mean pressure difference caused by combined centrifugal and swirling effects. The variance of the pressure fluctuations exhibits a wide region with negative logarithmic decay with the wall distance, due to the presence of a hierarchy of wall-attached eddies, which we clearly trace in spectral maps, and which are generally isotropic in nature. On the other hand, the largest eddies are strongly non-isotropic, and are mainly elongated in the azimuthal direction. POD is used to prove self-similarity of the attached eddies, whose typical scales are found to be linearly proportional to the wall distance of their center.

Experimental investigation of expansion effect on shock wave boundary layer interaction near a compression ramp

The experiment is conducted to investigate the effect of expansion on the shock wave boundary layer interaction near a compression ramp. The small-angle expansion with an angle degree of 5° occurs at different positions in front of the compression ramp. The particle image velocimetry and flow visualization technology show the flow structures, velocity field, and velocity fluctuation near the compression ramp. The mean pressure distribution, pressure fluctuation, and power spectral density are measured by high-frequency response pressure transducers. The experimental results indicate that the expansion before the compression ramp position affects the shock wave boundary layer interaction to induce a large-scale separation. But the velocity fluctuation and pressure fluctuation are attenuated near the large-scale flow separation region. When the expansion occurs closer to the compression ramp, the expansion has a more significant impact on the flow. The fluctuation of velocity and pressure is significantly attenuated, and the wall pressure rise of the separation point is reduced obviously. And the characteristic low-frequency spectrum signal related to the unsteadiness of the shock wave boundary layer interaction is significantly suppressed. In addition, variation of the separation region scale at different compression angle degrees is distinctive with the effect of expansion.

An atomistic approach to study the dynamic and structural response in 2D nanofiller reinforced polyethylene nanocomposites under ultra-short shock pulse loading

An atomistic model was developed to systematically investigate the shock propagation and its attenuation behavior in two-dimensional (2D) nanofiller reinforced polyethylene (PE) nanocomposites. Under the framework of non-equilibrium molecular dynamics, an explicit ultra-short shock pulse with an impact velocity of 1.0 km/s was applied on the mono- and bi-crystalline 2D nanofiller reinforced PE nanocomposites. Their dynamic and structural responses were captured in terms of particle velocity, mean pressure, kinetic energy, and radial distribution function (RDF). The trends depicted that the energy dissipated from graphene (GNS) and boron nitride nanosheet (BNNS) reinforced PE nanocomposites were 30.69% and 24.92%, respectively, whereas the energy dissipation improves up to 45.33% and 41.77% for bi-crystalline GNS and bi-crystalline BNNS, respectively. Higher energy dissipation in GNS reinforced PE nanocomposites can be attributed to extreme impedance mismatch and superior interfacial interaction energy between GNS and PE chains.

Turbulent flows along a streamwise external corner

A numerical investigation of turbulent flows in straight pipes with a cross-section in the form of a circular sector at different values of the apex angle${\alpha }$is carried out. Main attention is paid to the cases${\alpha }>{\rm \pi}$when the cross-section boundary contains a convex external corner. An analytical solution to the problem of laminar flow in the geometry under consideration is given. The law of resistance is determined, and velocity distributions over the pipe cross-section are obtained. The singularity of wall shear stress$\tau _w\sim r^{{\rm \pi} /{\alpha }-1}$at${\alpha }>{\rm \pi}$is shown which persists in turbulent flows as well. Turbulent flows are simulated at$Re=2000$and four apex-angle values${\alpha }=5{\rm \pi} /4$,$3{\rm \pi} /2$,$7{\rm \pi} /4$and$2{\rm \pi}$, as well as at a larger$Re=4000$and${\alpha }=3{\rm \pi} /2$. The characteristic features of the mean-velocity distributions are outlined, and their relationship with secondary flows is shown. The velocity of the secondary flow in the vicinity of the external corner reaches the value$6.3\,\%$of the bulk flow velocity at the maximum${\alpha }=2{\rm \pi}$. A characteristic feature of the considered flows is the presence of intense transverse velocity fluctuations in the region above the apex corner. It is hypothesized that the cause of these fluctuations may be the linear instability of the mean flow field. The studies of linear stability confirm this hypothesis. Considerable attention is paid to the study of secondary flows in the vicinity of the external corner. In general, the formation of these flows fits into the scheme proposed earlier (Nikitin,J. Fluid Mech., vol. 917, 2021, A24; Nikitinet al.,Fluid Dyn., vol. 56, issue 4, 2021, pp. 513–538), according to which the main mechanism is the centrifugal force arising from the fluctuating flow over the corner in the transverse plane. It is shown that the shape of secondary flows is reproduced to a certain extent by the vortical part of the Reynolds-stress force field. The latter may be found from the Helmholtz decomposition. The features of the mean pressure distribution associated with the secondary flow and with the presence of a region of concentrated fluctuations above the external corner are explained.

Investigation of the mean pressure field in the wing-wall junction region

A junction flow formed between a wing and a wall is ubiquitous in engineering applications and impacts aerodynamic performance and energy consumption. This paper studies the mean pressure field in the junction region around wall-mounted wings with NACA 0012 and NACA 6412 wing section profiles. Mean surface pressure measurements have been performed in the University of New South Wales large aerodynamic wind tunnel at a chord-based Reynolds number of 87,000 and a range of angles of attack spanning 0-12 degrees. Results of three-dimensional RANS simulations for both wing profiles using five turbulence models are also presented. The experimental results show that the wall significantly changes the mean pressure distribution around the wing, especially in the region upstream of the boundary layer reattachment point. The γ-Reθ SST turbulence model has a transition predictor for the boundary layer above the airfoil far away from the bottom wall and delivers the most accurate pressure field prediction in the wing-wall junction region. The junction pressure structure is well-captured except the LE stagnation region and the suction side of the trailing edge. The Reynolds Stress models, however, do not show any advantage in predicting the mean pressure field at high angles of attack.

An experimental study on flow over two finite wall-mounted square cylinders in a staggered arrangement

Wind tunnel measurement is conducted over two finite wall-mounted square cylinders in a staggered arrangement for various normalized cylinder distance (P* = Ps/d, where d is the cylinders width). For all experiments, two identical square cylinders with an aspect ratio of 7 are positioned on a line with a staggered angle of 45° relative to the incoming flow direction. The mean pressure distribution on the surfaces of the cylinders, aerodynamic force coefficients and Strouhal numbers are obtained at a Reynolds number Red = 6 × 104 for 1.42 ≤ P* ≤ 5.7. Two flow regimes namely single-body (P* ≤ 2.1) and modulated-periodic (2.8 ≤ P* ≤ 5.7) are identified. The flow is also visualized using the smoke methods at P* = 4 to provide further verification on the modulated-periodic flow patterns. Both cylinders for the single-body regime shed vortices with the same frequencies, whereas the downstream cylinder in the modulated-periodic one represents three dominant frequencies in the spectral analysis. It is also observed that the downstream cylinder experiences a higher drag for P* ≥ 4.2.

Two dimensional unsteady flow past a square cylinder: Influence of proximal plane wall and power-law index

A numerical analysis using Ansys Fluent has been done to investigate the flow physics involved in both Newtonian and Power-law (shear-thinning) fluid flow over a square cylinder, placed near a wall with altered gap-to-side (G/a) ratios (where G represents the gap between the cylinder and the plane wall and a represents side of the square cylinder). At a constant Reynolds number, Re = 150, numerical calculations for unsteady, incompressible Navier-Stroke (N-S) equation with varying viscosity is carried out by using second-order upwind momentum and least square cell based pressure solver. It is found that for small G/a ratio shear-thinning fluid exhibits vortex shedding which was absent for Newtonian fluid. G/a ratio and power-law index have a strong influence on the coefficient of lift and coefficient of drag (CL and CD respectively). With the increase in gap-ratio, CD decreases and CL increases at a constant value of power-law (n) index. Also, CL increases and CD reduces with a decrease in n at a constant value of G/a. Detailed observation of instantaneous and mean vorticity and streamlines, coefficient of lift and drag, mean pressure distribution over the cylinder and the plane wall and recirculation length have been presented.

Large-Eddy Simulations of Transverse Jet Mixing and Flame Stability in Supersonic Crossflow

Large-eddy simulations (LES) combined with the partially stirred reactor (PaSR) combustion model are employed to investigate the turbulent mixing, combustion mode, and flame stability of a sonic hydrogen jet injecting into high-enthalpy supersonic crossflow at three momentum flux ratios , in other words, 0.71, 2.11, and 4.00, respectively. The LES accuracy in terms of the turbulent kinetic energy, power spectra density, and subgrid Damköhler number is carefully addressed against various LES resolution criteria and the experimental mean pressure distribution on the upper wall. The ignition processes with autoignition and shock compression effects are identified and analyzed. At , the shock-induced ignition occurs behind the reflected shock wave, and the combustion heat release is dominated by the premixed combustion. While for the high jet-to-crossflow momentum flux ratios, for example, and 4.00, ignition happens around the jet orifice due to the strong bow shock and reflected shock effects, and the combustion heat releases are dominated by the nonpremixed combustion. Furthermore, the mechanisms of flame stabilization, local extinction, and reignition in the transverse jet combustion in supersonic crossflow are further analyzed with the chemical explosive mode analysis.

Numerical study of flow characteristics around rectangular cylinders in tandem

The phenomena of mean pressure distribution around two square cylinders in tandem has been carried out. Also the mean velocity distribution in the wake of the two cylinder are studies. The Navier-Stokes equation for conservation of momentum and the continuity equation for conservation of mass are used to study the pressure and velocity distribution for various longitudinal spacing of the cylinder. The turbulent k-ω stationary flow at a high Reynolds number of Re= 6.5e4 is used to simulate single-phase flow. A critical longitudinal spacing equal to four times the side dimension of the square cylinder (L/D= 4) is investigated. Beyond that critical spacing both the cylinders are subjected to drag force. When L/D < 4 only the downstream cylinder is subjected to negative drag. Also, the form of the streamline profile behind upstream cylinder is different from that of single cylinder. However, beyond L/D= 4 the nature of the pressure behind both of the cylinder becomes kind of like that of the single cylinder.

Numerical Investigation of Supersonic Turbulent Flow Over Open Cavities

The study of cavity flow is one of the most important research topics of unsteady aerodynamics. Supersonic turbulent flows over a cavity are mostly encountered in missiles, turbomachinery, and high-speed aircraft. The turbulence inside the cavity gives rise to excess drag, acoustic waves, pressure fluctuations, and vibrations which may lead to numerous problems like excess fuel consumption, failure of missile trajectory or mechanical parts, and aerodynamic heating. We conduct numerical simulations to investigate the flow and compare our results to existing experimental data to show quantitative validation. We then investigate the effect of the Mach number of the turbulent supersonic flow on the pressure contours and the vortical structures inside the cavity and the subsequent effect on other flow parameters like acoustic waves. Thereafter, we propose two modifications of the cavity geometry, a) slanted edges and b) smoothened corners with slanted edges, to improve the aerodynamic performance. ANSYS ICEM tool has been used for the fine mesh generation of our cavity geometry and all the simulation was run on ANSYS fluent software. The K omega SST turbulence model was used for the simulation as it can capture the near wall property with reasonable accuracy and a grid independence study was carried out to find the correct solution of the Navier Stokes equation was proceeding by the solver. The parameters like pressure contours, streamline pattern, coefficient of pressure distribution and sound pressure level (SPL) has been found and compared for both modified and unmodified open cavity at different Mach number. The modifications suggested show a significant improvement over the open cavity designs in the mean pressure distribution and sound pressure level distributions.

Reynolds number effects on the wind pressure distribution on spherical storage tanks

Spherical storage tanks are important industrial equipment with typical bluff body shape, the aerodynamic loads of which are significantly affected by the Reynolds number (Re). In the present research, wind tunnel tests are carried out on spherical storage tanks with 4–20 supporting columns and a range of surface roughness (ks/D ​= ​0, 0.002 and 0.004) under uniform and atmospheric boundary layer (ABL) flows with Re ranging from 105 to 106. The Reynolds number effects on the aerodynamic drag and uplift coefficients, mean and RMS wind pressure distributions are analyzed. The separation angle on the upper hemi-sphere shifts from 135° (super-critical) to 120° (trans-critical) at Re ​≈ ​6.5 ​× ​105 for a smooth tank in the uniform flow. The trans-critical self-modeling Re range criteria are determined by analyzing the similarity of mean and RMS wind pressure coefficient distribution patterns. It is indicated that in the ABL flow, when ks/D ​≥ ​0.002, the mean and RMS wind pressure distribution of a spherical tank in the trans-critical self-modeling Re range can be simulated with merely Re ​> ​2.0 ​× ​105, thus providing a guidance for wind tunnel simulation of sphere-like bodies. Moreover, a piecewise mean wind pressure coefficient model in the trans-critical self-modeling Re range for spherical storage tanks is established for engineering reference.

On the three-dimensional flow development around circular finned cylinders

The three-dimensional flow development around the circular finned cylinders is investigated numerically. Three finned cylinders with constant fin pitch (p), fin thickness (t), and effective diameter (Deff) and a range of diameter ratios (Df/Dr) within 1.25 ≤ Df/Dr ≤ 2.5 are considered in this study. One bare cylinder with a diameter equivalent to the effective diameter of the finned cylinders is also considered. The numerical simulations are performed using the large eddy simulation turbulence model for a free-stream velocity corresponding to the Reynolds number Re = 3900, defined based on the effective diameter (Deff). This study provides novel insights into the flow physics in the channel between the fins and its effect on the three-dimensional flow development. The results accentuate that the three-dimensional flow development around the finned cylinders fundamentally differs from that of the bare cylinder. In particular, the flow separation topology at the surface of the finned cylinders differs significantly from that of the bare cylinder due to flow entrainment between the fins. The distinct flow separation leads to the formation of streamwise edge vortices, which induces downwash flow in the near wake of the finned cylinders. The combined effect of the entrainment between the fins and downwash flow affects the vortex formation length and mean pressure distribution around the cylinder. As a result, the structural loading on the surface of the finned cylinders is distinctively different from that of the bare cylinder with profound dependence on the fin parameters.

Effects of corner modifications on wind loads and local pressures on walls of tall buildings

In this study, the aerodynamic characteristics of tall buildings with corner modifications (e.g., local wind force coefficients, mean pressure distributions, normalized power spectrum density, and extreme local pressure) were examined. Wind tunnel experiments were conducted to measure the wind pressures on building models with different heights and recessed corners of different ratios. At a wind direction of a = 0° (i.e., wind blowing on the front of a building), corner modifications effectively reduced wind forces in all cases. Specifically, small corner modification ratios reduced wind forces more effectively than their larger counterparts. However, corner modifications resulted in extreme local pressure on building surfaces. In addition, for small corner modification ratios, the probability of extreme local pressure occurring at a = 0° was high. This probability was also high for large corner modification ratios at a = 15° (i.e., wind blowing slightly obliquely on the front of a building) because wind blowing obliquely creates substantial vortex shedding on one side surface and extreme negative pressure over one building side surface. Results of computational fluid dynamic modeling were adopted to determine details of the aerodynamic characteristics of tall buildings with corner modifications.

Numerical investigation of transonic buffet on supercritical airfoil considering uncertainties in wind tunnel testing

To improve the predictive ability of computational fluid dynamics (CFD) on the transonic buffet phenomenon, NASA SC(2)-0714 supercritical airfoil is numerically investigated by noninstructive probabilistic collocation method for uncertainty quantification. Distributions of uncertain parameters are established according to the NASA wind tunnel report. The effects of the uncertainties on lift, drag, mean pressure and root-mean square pressure are discussed. To represent the stochastic solution, the mean and standard deviation of variation of flow quantities such as lift and drag coefficients are computed. Furthermore, mean pressure distribution and root-mean square pressure distribution from the upper surface are displayed with uncertainty bounds containing 95% of all possible values. It is shown that the most sensitive part of flow to uncertain parameters is near the shock wave motion region. Comparing uncertainty bounds with experimental data, numerical results are reliable to predict the reduced frequency and mean pressure distribution. However, for root-mean square pressure distribution, numerical results are higher than the experimental data in the trailing edge region.