An experimental investigation of flow-induced acoustic resonance and flow field in a closed side branch system using a high time-resolved PIV technique

Abstract  

Systems with closed side branches are liable to an excitation of sound known as cavity tone. It may occur in pipe branches leading to safety valves or to boiler relief valves. The outbreak mechanism of the cavity tone has been ascertained by phase-averaged pressure measurements in previous research, while the relation between sound propagation and the flow field is still unclear due to the difficulty of detecting the instantaneous velocity field. It is possible to detect the two-dimensional instantaneous velocity field using high time-resolved particle image velocimetry (PIV). In this study, flow-induced acoustic resonance in a piping system containing closed side branches was investigated experimentally. A high time-resolved PIV technique was used to measure the gas flow in a cavity. Airflow containing oil mist as tracer particles was measured using a high-frequency pulse laser and a high-speed camera. The present investigation on the coaxial closed side branches is the first rudimentary study to visualize the fluid flow two-dimensionally in a cross-section using high time-resolved PIV, and to measure the pressure at the downstream side opening of the cavity by microphone. The fluid flows at different points in the cavity interact, with some phase differences between them, and the relation between the fluid flows was clarified.     

Stereoscopic PIV measurements of flow in and around axial flow fan

To analyze the complex three-dimensional flow structure of an axial flow fan and determine the validity of its application, PIV is used to provide detailed space and time resolved experimental data for understanding and control of flow field. The high resolution stereoscopic PIV system was successfully employed in this study for the investigation of flow structure around the axial flow fan. Using the once-per-revolution signal from the rotor, image fields were captured at a fixed position of the blades and hence provides the ability to do phase-averaging. The three-dimensional instantaneous velocity fields, phase-averaged velocity fields, instantaneous and mean vorticity distributions of the stereoscopic PIV measurement results were represented at typical planes of the flow field. Phaseaveraged velocity fields were calculated based on 200 frames of the instantaneous stereoscopic PIV measurement results. From the velocity distribution, the vorticity and turbulent intensity distribution, which are known to be major factors of fan noise, were calculated and its diffusion was discussed as they travel downstream. From the reconstructed three-dimensional velocity iso-surface at 8 cross planes of the outlet flow fields, the three-dimensional features can be seen clearly.     

Planar velocity visualization in high-speed wedge flow using Doppler Picture Velocimetry (DPV) compared with Particle Image Velocimetry (PIV)

A technique for visualizing a velocity field in an entire plane has been developed by taking ‘Doppler Pictures’ using Michelson interferometry. With the Doppler Picture Velocimetry (DPV), information about the instantaneous and local velocities of tracers passing through a light sheet is available. The technique for taking and processing the Doppler pictures has been improved recently and the state-of-the-art of the DPV method will be described with an application in high-speed fluid flows showing the velocity distribution in a light sheet plane crossing a supersonic wedge flow generated in the high-energy shock tunnel STB of ISL. A comparison with Particle Image Velocimetry (PIV) velocity visualizations is also presented.     

Application of DPIV to study both steady state and transient turbomachinery flows

Digital particle imaging velocimetry (DPIV) is a powerful measurement technique, which can be used as an alternative or complementary approach to laser doppler velocimetry (LDV) in a wide range of research applications. The instantaneous planar velocity measurements obtained with PIV make it an attractive technique for use in the study of the complex flow fields encountered in turbomachinery. The planar nature of the technique also significantly reduces the facility run time over point-based techniques. Techniques for optical access, light sheet delivery, CCD camera technology and particulate seeding are discussed. Results from the successful application of the PIV technique to both the blade passage region of a transonic axial compressor and the diffuser region of a high speed centrifugal compressor are presented. Both instantaneous and time-averaged flow fields were obtained. The averaged flow field measurements are used to estimate the flow turbulence intensity. The instantaneous velocity vector maps obtained during compressor surge provide previously unobtainable insight into the complex flow field characteristics occurring during short lived surge events. These flow field maps illustrate the true power of the DPIV technique.     

Investigation on the flow characteristics inside an automotive HVAC system with varying ventilation mode

The ventilation flow in a heating, ventilation and air conditioning (HVAC) module of a passenger car was investigated experimentally. Three different ventilation modes with varying temperature mode were tested to study the effect of ventilation mode on the velocity field inside the HVAC module. For each mode, more than 450 instantaneous velocity fields were measured using a particle image velocimetry (PIV) velocity field measurement technique. The instantaneous velocity fields were ensemble averaged to obtain the spatial distribution of mean velocity and spanwise vorticity. The present work highlights the usefulness of the PIV technique for the analysis of the flow inside an HVAC module. The experimental results can be used not only to understand and improve the ventilation flow of an HVAC module but also to validate numerical predictions.     

Volume-resolved flame chemiluminescence and laser-induced fluorescence imaging

There is significant need for optical diagnostic techniques to measure instantaneous volumetric vector and scalar distributions in fluid flows and combustion processes. This is especially true for investigations where only limited optical access is available, such as in internal combustion engines, furnaces, flow reactors, etc. While techniques such as tomographic PIV for velocity measurement have emerged and reached a good level of maturity, instantaneous 3D measurements of scalar quantities are not available at the same level. Recently, developments in light field technology have progressed to a degree where implementation into scientific 3D imaging becomes feasible. Others have already demonstrated the utility of light field technology toward imaging high-contrast particles for PIV and for imaging flames when treated as single-surface objects. Here, the applicability and shortcomings of current commercially available light field technology toward volumetric imaging of translucent scalar distributions and flames are investigated. Results are presented from imaging canonical chemiluminescent and laser-induced fluorescent systems. While the current light field technology is able to qualitatively determine the position of surfaces by locating high-contrast features, the correlation-based reconstruction algorithm is unable to fully reconstruct the imaged objects for quantitative diagnostics. Current analysis algorithms are based on high-contrast correlation schemes, and new tools, possibly based on tomographic concepts, will have to be implemented to reconstruct the full 3D structure of translucent objects for quantitative analysis.     

Whole field spatial frequency analysis of double-or multiple-exposed PIV images

Speckle Velocimetry or high-image density PIV gives a velocity vector map of a twodimensional flow field by point-by-point spatial frequency analysis of local pattern at an interrogation spot in a double- or a multiple-exposed image of laser speckle or pseudo-speckle pattern generated by pulsed laser-light-sheet illuminations of a plane in the flow densely seeded with fine particles. The whole field spatial frequency analysis of the double- or multiple-exposed PIV image gives more useful information of the flow field. Optical Fourier transform is a conventional technique not only for the local spatial frequency analysis but for the whole field analysis. Filtering of spatial frequency is one of the typical techniques for the latter which can reconstruct a velocity contour or a component velocity contour map of the flow. Fundamentally this technique is a simple and efficient analogue method to get more information in the velocity field analysis of the flow than a digital image processing, but in practical applications the measurement is restricted to simple flows because of optical noise and low spatial resolution. In order to improve the technique, the fundamental characteristics of the filtering and the noise yielded in the filtering process must be investigated. Meanwhile, wavelet analysis can also be applied to the whole field spatial frequency analysis of PIV image. In this paper the filtering technique is examined by numerical convolution integral, and the results obtained are compared with ones obtained by the wavelet analysis.     

3D imaging of evaporating fuel droplets by stereoscopic PIV

A gun-type burner is a widely used oil burner for industrial and domestic applications. The oil is pressure-atomized and mixed with air generating a recirculating, swirling flow. Because of the surrounding flame, fuel droplets evaporate, being difficult to obtain information on droplets’ dynamics. Several laser techniques have been applied to this burner for spray diagnosis. PDA provides information about droplet size and velocity but can say little about the instantaneous spatial structures in the flow. Planar laser techniques as PIV can describe the 2D instantaneous spatial structures, but cannot provide information about the 3D structures in the flow. Then Stereoscopic PIV was applied. This technique allows us to measure the full 3D velocity vector map in a whole fluid plane. This paper has a double purpose. Firstly, to visualize the 3D structures which are present in the burner; secondly, to show that Stereoscopic PIV is an applicable technique for the diagnosis of an evaporating spray.     

Schlieren “PIV” for turbulent flows

The possibility of using commercial PIV equipment combined with schlieren optics to measure the velocity fields of turbulent flows is explored. Given a sufficiently high Reynolds number and adequate refractive flow differences, turbulent eddies can serve as the PIV “particles” in a schlieren image or shadowgram. The PIV software analyzes motion between consecutive schlieren or shadowgraph frames to obtain velocity fields. Velocimetry examples of an axisymmetric sonic helium jet in air and a 2D turbulent boundary layer at Mach 3 are shown. Due to optical path integration, axisymmetric flows require the inverse Abel transform to extract center-plane velocity data. Conditions for optimum schlieren sensitivity are examined. In its present embodiment, “schlieren PIV” is not useful for laminar flows nor for fully 3D flows. Otherwise it functions much like standard PIV under conditions where individual particles are not resolved and velocimetry is instead based on correlation of the motion of turbulent structures. “Schlieren PIV” shows significant promise for general refractive turbulent flow velocimetry if its integrative nature can be overcome through sharp-focusing optics.     

Analysis of Two-Phase Cavitating Flow with Two-Fluid Model Using Integrated Boltzmann Equations

In the present work, both computational and experimental methods are employed to study the two-phase flow occurring in a model pump sump. The two-fluid model of the two-phase flow has been applied to the simulation of the three-dimensional cavitating flow. The governing equations of the two-phase cavitating flow are derived from the kinetic theory based on the Boltzmann equation. The isotropic RNG$k-\epsilon-k_{ca}$ turbulence model of two-phase flows in the form of cavity number instead of the form of cavity phase volume fraction is developed. The RNG $k-\epsilon-k_{ca}$ turbulence model, that is the RNG$k-\epsilon$ turbulence model for the liquid phase combined with the $k_{ca}$model for the cavity phase, is employed to close the governing turbulent equations of the two-phase flow. The computation of the cavitating flow through a model pump sump has been carried out with this model in three-dimensional spaces. The calculated results have been compared with the data of the PIV experiment. Good qualitative agreement has been achieved which exhibits the reliability of the numerical simulation model.     

Laser flow visualization and velocity fields by particle image velocimetry in an electrostatic precipitator model

Although improving electrostatic precipitator (ESP) collection of fine particles (micron and submicron sizes) remains of interest, it is not yet clear whether the turbulent flow patterns caused by the presence of electric field and charge in ESPs advance or deteriorate fine particle precipitation process. In this paper, results of the laser flow visualization and Particle Image Velocimetry (PIV) measurements of the particle flow velocity fields in a wire-to-plate type ESP model with seven wire electrodes are presented. Both experiments were carried out for negative and positive polarity of the wire electrodes. The laser flow visualization and PIV measurements clearly confirmed formation of the secondary flow (velocity of several tens of cm/s) in the ESP model, which interacts with the primary flow. The particle flow pattern changes caused by the strong interaction between the primary and secondary flows are more pronounced for higher operating voltages (higher electrohydrodynamic numbernehd) and lower primary flow velocities (lower Reynolds number Re). The particle flow patterns for the positive voltage polarity of the wire electrodes are more stable and regular than those for the negative voltage polarity due to the nonuniformity of the negative corona along the wire electrodes (tufts).     

Clustering of the low-inertia particle number density field in random divergence-free hydrodynamic flows

We consider the diffusion of the low-inertia particle number density field in random divergence-free hydrodynamic flows. The principal feature of this diffusion is the divergence of the particle velocity field, which results in clustering of the particle number density field. This phenomenon is coherent, occurs with a unit probability, and must show up in almost all realizations of the process dynamics. We calculate the statistical parameters that characterize clustering in three-dimensional and two-dimensional random fluid flows and in a rapidly rotating two-dimensional random flow. In the former case, the vortex component of the random divergence-free flow generates a vortex component of the low-inertia particle velocity field, which generates a potential component of the velocity field through advection. By contrast, in the case of rapid rotation, a potential component of the velocity field is generated directly by the vortex component of the random divergence-free flow (linear problem).     

PIV measurements of the flow and turbulent characteristics of a round jet in crossflow

The instantaneous and ensemble averaged flow characteristics of a round jet issuing normally into a crossflow was studied using a flow visualization technique and Particle Image Velocimetry measurements. Experiments were performed at a jet-to-crossflow velocity ratio, 3.3 and two Reynolds numbers, 1,050 and 2,100, based on crossflow velocity and jet diameter. Instantaneous laser tomographic images of the vertical center plane of the crossflow jet show that there exists very different natures in the flow structures of the near field jet due to Reynolds number effect even though the velocity ratio is same. It is found that the shear layer becomes much thicker when the Reynolds number is 2,100 because of the strong entrainment of the inviscid fluid by turbulent interaction between the jet and crossflow. The mean and second order statistics are calculated by ensemble averaging over 1,000 realizations of instantaneous velocity fields. The detail characteristics of mean flow field, streamwise and vertical rms velocity fluctuations, and Reynolds shear stress distributions are presented. The new PIV results are compared with those from previous experimental and LES studies.     

Dynamic PIV measurement of a high-speed flow issuing from vent-holes of a curtain-type airbag

A curtain-type airbag is a safety device designed to protect passengers from the side collisions of a car. The curtain-type airbag system consists of an inflator, a fill-hose, and a curtain-airbag. The fill-hose is a passageway and distributor of the exploded gases from the inflator to the airbag through vent-holes. Although the design of vent-holes is important for proper deployment of the airbag, it is very difficult to measure the exceedingly high speed flow issuing from the vent-holes by using conventional measurement methods. In this study, we employed a dynamic PIV technique to measure the temporal evolution of instantaneous velocity fields of the flow ejecting from the vent-holes. From the velocity field data measured at a frame rate of 2000 fps, the temporal variation of the volume flux from vent-holes was also evaluated for the diagnosis of airbag performance. The flows ejecting from the vent-holes showed high velocity fluctuations, and the maximum velocity was about 480 m/s. The instantaneous velocity fields in the initial stage showed a swaying motion of a high-speed jet. The accumulated volume flux from the vent-holes was also compared at each vent-hole region.     

基于相场方法的孔隙尺度油水两相流体流动模拟

基于真实岩心颗粒粒径分布,利用过程法构建疏松砂岩油藏的三维孔隙结构模型,利用相场方法建立两相流体流动数学模型并利用有限元方法进行求解,研究驱替速度、流体性质、润湿性对剩余油分布以及采出程度的影响.结果表明：驱替速度的增大和油水粘度比的减小会导致较大的毛管数,进而有利于采出程度的提高;就润湿性而言,水湿条件下毛管力是水驱油的动力,而在油湿条件下是阻力,因此水湿岩心采出程度更高.同时,从孔隙尺度对油水渗流机理及剩余油分布机理进行揭示,结果表明：由于多孔介质的复杂孔隙结构,流体在流经不同孔隙时呈现不同的流动特征,进而对油水两相流整体的压力分布、流速分布造成重要影响.     

Direct measurement of fluid velocity gradients at a wall by PIV image processing with stereo reconstruction

Velocity gradient is typically estimated in Particle Image Velocimetry (PIV) by differentiating a measured velocity field, which amplifies noise in the measured velocities. If gradients near a boundary are sought, such noise is usually greater than in bulk fluid, because of small tracer displacement, uncertainty in the effective positions of velocity vectors, intense deformation of tracer patterns, and laser reflection. We consider here a modified form of the Particle Image Distortion (PID) method todirectly calculate velocity gradients at a fixed wall, and refer it as “PIV/IG” (“Interface Gradiometry”). Results from synthetic 2D PIV images suggest our method achieves higher SNR and accuracy than velocity differentiation. Also, we have developed a procedure to reconstruct three-dimensional velocity gradient at a fixed wall the two non-zero components from PIV/IG data obtained in stereo views; these equations simplify considerably thanks to the no-slip condition. Experimental data from the bottom wall of turbulent open channel flow appear to suffer from a form of pixel locking. As with standard PIV, this underlines the importance of adequate tracer diameter in the images, sufficient seeding density, and of dynamic range of the camera sensor.     

Calculating pressure fields on the basis of PIV-measurements for supersonic flows

The velocity fields obtained by PIV (Particle Image Velocimetry) in supersonic flows are not sufficient to determine the integral characteristics of the flow. Additional data, for example, on pressure can be obtained from the solution of the Navier?Stokes equations. For incompressible flows, the solution of these equations is not too complicated. However, for supersonic flows, the need to take into account the flow density and the increasing number of experimental errors make it more difficult. This paper proposes a new method for calculating density and pressure from PIV data on the basis of the continuity equation. This method is robust and easy to implement for compressible flows.     

Holographic flow visualization as a tool for studying three-dimensional coherent structures and instabilities

Holography is capable of three-dimensional (3D) representation of spatial objects such as fluid interfaces and particle ensembles. Based on this, we adapt it into a 3D flow visualization tool called Holographic Flow Visualization (HFV). This technique provides a novel means of studying spatially and temporally evolving complex fluid flow structures marked by a disperse phase or interfaces of different fluids. This paper demonstrates that HFV is a straightforward technique, especially when the In-line Recording Off-axis Viewing (IROV) configuration is used. The technique can be applied either as a stand-alone experimental tool for studying scalar-based coherent structures, flow instabilities, interactions of different fluids driven by fluid dynamics, interfacial phenomena, or as a precursor to volumetric 3D velocity vector field measurement of complex transient flow dynamics. Experimental results in several complex fluid flows and flames demonstrate the effectiveness of HFV. Different methods are used to mark flow structures undergoing different instabilities: 1) a vortex ring grown out of a drop of polymer suspension falling in water, 2) cascade of a bag-shaped drop of milk in water, and 3) internal flow structures of a jet diffusion flame.     

Numerical simulation of instantaneous flow structure of swirling and non-swirling coaxial-jet particle-laden turbulence flows

A subgrid scale two-phase second-order-moment (SGS-SOM) model based on the two-fluid continuum approach is presented for the analysis of the instantaneous flow structures of swirling and non-swirling coaxial-jet particle-laden turbulence flows. Since the interaction between the two-phase subgrid scale stresses and the anisotropy of two-phase subgrid scale stresses is fully considered, it is superior to the conventional subgrid scale model on the basis of single gas phase or together with their similar forms for the particle phase for not taken these characters thoroughly into account. The swirling numbers s=0.47 and s=0 of coaxial-jet particle-laden turbulence flows (measured by M. Sommerfeld, H.H. Qiu, Detailed measurements in a swirling particulate two-phase flow by a phase Doppler anemometer, Int. J. Heat Fluid Flow 12 (1991) 20-28) are numerically simulated by large eddy simulation using this model, together with a Reynolds-averaged Navier-Stokes model using the unified second-order-moment two-phase turbulence model (RANS-USM). The instantaneous results show that the multiple recirculating gas flow structure is similar to that of single-phase swirling flows; but the particle flow structure contains less vortices. Both SGS-SOM and RANS-USM predicted that the two-phase time-averaged velocities and the root-mean-square fluctuation velocities are validated and are in good agreement with the experimental results. It is seen that for the two-phase time-averaged velocities both the models give almost the same results, hence the RANS-USM modeling is validated by large eddy simulation. For the two-phase root-mean-square fluctuation velocities the SGS-SOM results are obviously better than the RANS-USM results.     

Unsteady flow visualization at part-load conditions of a radial diffuser pump: by PIV and CFD

The present study provides flow visualization on complex internal flows in a radial diffuser pump under part-load conditions by using the three-dimensional Navier-Stokes code CFX-10 with Detached Eddy Simulation (DES) turbulence model. Particle Image Velocimetry (PIV) measurements have been conducted to validate numerical results. The CFD results show good agreements with experimental ones on both the phase-averaged velocity fields and turbulence field. The detailed flow analysis shows that no separation occurs at 0.75Q_des although a low-velocity zone develops on the rear impeller suction side. Steady flow separations are observed on the impeller suction sides at 0.5Qdes but with different onsets and amounts. When reducing the flow rate to 0.25Q_des, CFD predicts different types of back flows in the impeller region, including steady leading edge separations, rotating vortex in the impeller wake region, and back flow on the impeller pressure side.