Endoscopic PIV measurements in a low pressure turbine rig

Particle-Image-Velocimetry (PIV) is a useful way to acquire information about the flow in turbomachinery. Several premises have to be fulfilled to achieve high-quality data, for example, optical access, low vibrations and low reflections. However, not all test facilities comply with these requirements. If there is no optical access to the test area, measurements cannot be performed. The use of borescopic optics is a possible solution to this issue, as the access required is very small. Several different techniques can be used to measure the three components of the velocity vector, one of which is Stereo-PIV. These techniques require either large optical access from several viewing angles or highly complex setups. Orthogonal light sheet orientations in combination with borescopic optics using Planar-PIV can deliver sufficient information about the flow. This study will show the feasibility of such an approach in an enclosed test area, such as the interblade space in a Low-Pressure-Turbine-Rig. The results from PIV will be compared with data collected with conventional techniques, such as the Five-Hole-Probe and the 2-component Hot-Wire-Anemometry. An analysis of time- and phase-averaged data will be performed.     

PIV, 2D-LIF and 1D-Raman measurements of flow field,composition and temperature in premixed gas turbine flames

Several laser diagnostic measurement techniques have been applied to study the lean premixed natural gas/air flames of an industrial swirl burner. This was made possible by equipping the burner with an optical combustion chamber that was installed in the high-pressure test rig facility at the DLR Institute of Combustion Technology in Stuttgart. The burner was operated with preheated air at various operating conditions with pressures up to p = 6 bar and a maximum thermal power of P = 1 MW.The instantaneous planar flow field inside the combustor was studied with particle image velocimetry (PIV). Planar laser induced fluorescence (PLIF) of OH radicals on a single-shot basis was used to determine the shape and the location of the flame front as well as the spatial distribution of reaction products. 1D laser Raman spectroscopy was successfully applied for the measurement of the temperature and the concentration of major species under realistic gas turbine conditions.Results of the flow field analysis show the shape and the size of the main flow regimes: the inflow region, the inner and the outer recirculation zone. The highly turbulent flow field of the inner shear layer is found to be dominated by small and medium sized vortices. High RMS fluctuations of the flow velocity in the exhaust gas indicate the existence of a rotating exhaust gas swirl. From the PLIF images it is seen that the primary reactions happened in the shear layers between inflow and the recirculation zones and that the appearance of the reaction zones changed with flame parameters. The results of the multiscalar Raman measurements show a strong variation of the local mixture fraction allowing conclusions to be drawn about the premix quality. Furthermore, mixing effects of unburnt fuel and air with fully reacted combustion products are studied giving insights into the processes of the turbulence–chemistry interaction.     

Three-component velocity field measurements of propeller wake using a stereoscopic PIV technique

A stereoscopic PIV (Particle Image Velocimetry) technique was used to measure the three-dimensional flow structure of the turbulent wake behind a marine propeller with five blades. The out-of-plane velocity component was determined using two CCD cameras with an angular displacement configuration. Four hundred instantaneous velocity fields were measured for each of four different blade phases, and ensemble averaged in order to find the spatial evolution of the propeller wake in the region from the trailing edge up to one propeller diameter (D) downstream. The influence of propeller loading conditions on the wake structure was also investigated by measuring the velocity fields at three advance ratios (J=0.59, 0.72 and 0.88). The phase-averaged velocity fields revealed that a viscous wake formed by the boundary layers developed along the blade surfaces. Tip vortices were generated periodically and the slipstream contracted in the near-wake region. The out-of-plane velocity component and strain rate had large values at the locations of the tip and trailing vortices. As the flow moved downstream, the turbulence intensity, the strength of the tip vortices, and the magnitude of the out-of-plane velocity component at trailing vortices all decreased due to effects such as viscous dissipation, turbulence diffusion, and blade-to-blade interaction.     

Physics-constrained bayesian neural network for fluid flow reconstruction with sparse and noisy data

In many applications, flow measurements are usually sparse and possibly noisy. The reconstruction of a high-resolution flow field from limited and imperfect flow information is significant yet challenging. In this work, we propose an innovative physics-constrained Bayesian deep learning approach to reconstruct flow fields from sparse, noisy velocity data, where equationbased constraints are imposed through the likelihood function and uncertainty of the reconstructed flow can be estimated. Specifically, a Bayesian deep neural network is trained on sparse measurement data to capture the flow field. In the meantime, the violation of physical laws will be penalized on a large number of spatiotemporal points where measurements are not available. A non-parametric variational inference approach is applied to enable efficient physicsconstrained Bayesian learning. Several test cases on idealized vascular flows with synthetic measurement data are studied to demonstrate the merit of the proposed method.     

Effect of gas expansion on the velocity of a Taylor bubble: PIV measurements

The effect of gas expansion on the velocity of a Taylor bubble was studied experimentally. The velocity field in the liquid ahead of a Taylor bubble was measured by particle image velocimetry (PIV), and the bubble velocity was measured with two pairs of laser diodes and photocells. The experiments were done in a 7.0 m long vertical tube with a 32 mm internal diameter. Solutions of carboxymethylcellulose (CMC) polymer with weight percentages between 0.01% and 0.1% were used. The expansion of slug gas induces an increase in the bubble velocity and a corresponding displacement of the liquid ahead of the bubble. The velocity of the bubble increases by an amount equal to the maximum velocity in the liquid displaced. For the solutions studied, the induced velocity profile was parabolic and the bubble velocity increase was equal to the liquid velocity at the tube axis, i.e., twice the mean velocity in the liquid displaced. The corrected velocity obtained by subtracting the velocity increase from the value of the bubble velocity is independent of the bubble length.     

Assessment of dual plane PIV measurements in wall turbulence using DNS data

Experimental dual plane particle image velocimetry (PIV) data are assessed using direct numerical simulation (DNS) data of a similar flow with the aim of studying the effect of averaging within the interrogation window. The primary reason for the use of dual plane PIV is that the entire velocity gradient tensor and hence the full vorticity vector can be obtained. One limitation of PIV is the limit on dynamic range, while DNS is typically limited by the Reynolds number of the flow. In this study, the DNS data are resolved more finely than the PIV data, and an averaging scheme is implemented on the DNS data of similar Reynolds number to compare the effects of averaging inherent to the present PIV technique. The effects of averaging on the RMS values of the velocity and vorticity are analyzed in order to estimate the percentage of turbulence intensity and enstrophy captured for a given PIV resolution in turbulent boundary layers. The focus is also to identify vortex core angle distributions, for which the two-dimensional and three-dimensional swirl strengths are used. The studies are performed in the logarithmic region of a turbulent boundary layer at z ⁺ = 110 from the wall. The dual plane PIV data are measured in a zero pressure gradient flow over a flat plate at Re _τ = 1,160, while the DNS data are extracted from a channel flow at Re _τ = 934. Representative plots at various wall-normal locations for the RMS values of velocity and vorticity indicate the attenuation of the variance with increasing filter size. Further, the effect of averaging on the vortex core angle statistics is negligible when compared with the raw DNS data. These results indicate that the present PIV technique is an accurate and reliable method for the purposes of statistical analysis and identification of vortex structures.     

A novel PIV technique for measurements in multiphase flows and its application to two-phase bubbly flows

A new experimental procedure for performing simultaneous, phase-separated velocity measurements in two-phase flows is introduced. Basically, the novel particle image velocimetry (PIV) technique is a combination of the three most often used PIV techniques in multiphase flows: PIV with fluorescent tracer particles, shadowgraphy, and the digital phase separation with a masking technique. The combination of these three independent measurement techniques is achieved by shifting the background intensity of a PIV recording to a higher, but uniform gray value level. In order to combine the advantages of these multiphase-PIV methods, a new PIV set-up was developed. With this set-up the velocity distributions of the two phases are measured simultaneously with only one b/w camera. This experimental set-up is aimed at providing a means for characterizing the modification of turbulence in the liquid phase by bubbles. This phenomenon is often called "pseudo-turbulence".     

Spatial resolution of a chemiluminescence sensor for local heat-release rate and equivalence ratio measurements in a model gas turbine combustor

The spatial resolution of a Chemiluminescence Sensor, based on focused Cassegrain optics, to detect the location of the reaction zone and heat-release rate in a model gas turbine combustor is reported. The sensor measures simultaneously the chemiluminescent intensities from OH* and CH* excited radicals in flames in order to obtain information on the local flame characteristics. The spatial resolution was evaluated by a combined theoretical and experimental study in laminar and turbulent flames and was supported by detailed chemistry calculations, including the chemiluminescent species, of unstrained one-dimensional flames. The experimental study involved simultaneous measurements of chemiluminescence with the sensor and laser-based reaction rate imaging, using the product of OH and CH₂O radicals obtained from planar laser-induced fluorescence (PLIF), and OH PLIF for the location of the reaction zone. The study quantified the influence of flame shape and dimensions and the direction of traverse of the focal region of the sensor through the flames on the spatial resolution, thereby identifying the limitations and optimising the applicability of the sensor. The sensor was used to obtain local time-dependent measurements of heat-release and equivalence ratio of a reacting mixture, based on the chemiluminescent intensity ratio of OH*/CH*, in a swirl-stabilised model gas turbine combustor and quantified the degree of air–fuel premixedness, probability of reaction and power spectra of pressure and chemiluminescent intensity fluctuations in two unsteady flames.     

Simultaneous PIV and concentration measurements in a gas-turbine combustor model

Simultaneous velocity and concentration measurements have been performed in a gas-turbine combustor model. Particle image velocimetry (PIV) was used to acquire planar velocity information and to identify coherent flow structures. The Mie scattering technique, based on a slightly modified experimental setup, was used for concentration measurements in this mixing flow. The degree of mixing was assessed by examining local concentration measurements while inhomogeneously seeding the primary and secondary stream of the mixing layer. Connections between flow field and concentration distribution were highlighted using the proper orthogonal decomposition algorithm (POD). Uncertainties and systematic errors for the PIV measurements due to the suboptimal seeding are discussed using a comparison with a second test series at optimal seeding conditions. Results are presented for several flow parameters and at various lateral planes.     

A multiphase,micro-scale PIV measurement technique for liquid film velocity measurements in annular two-phase flow

Prediction methods for two-phase annular flow require accurate knowledge of the velocity profile within the liquid film flowing at its perimeter as the gradients within this film influence to a large extent the overall transport processes within the entire channel. This film, however, is quite thin and variable and traditional velocimetry methods have met with only very limited success in providing velocity data. The present work describes the application of Particle Image Velocimetry (PIV) to the measurement of velocity fields in the annular liquid flow. Because the liquid is constrained to distances on the order of a millimeter or less, the technique employed here borrows strategies from micro-PIV, but micro-PIV studies do not typically encounter the challenges presented by annular flow, including very large velocity gradients, a free surface that varies in position from moment to moment, the presence of droplet impacts and the passage of waves that can be 10 times the average thickness of the base film. This technique combines the seeding and imaging typical to micro-PIV with a unique lighting and image processing approach to deal with the challenges of a continuously varying liquid film thickness and interface. Mean velocity data are presented for air–water in two-phase co-current upward flow in a rectangular duct, which are the first detailed velocity profiles obtained within the liquid film of upward vertical annular flow to the authors’ knowledge. The velocity data presented here do not distinguish between data from waves and data from the base film. The resulting velocity profiles are compared with the classical Law of the Wall turbulent boundary layer model and found to require a decreased turbulent diffusivity for the model to predict well. These results agree with hypotheses previously presented in the literature.     

Investigation of aerodynamic effects of coolant ejection at the trailing edge of a turbine blade model by PIV and pressure measurements

In order to simulate the thick trailing edges of turbine blades a slotted plate profile together with a newly designed nozzle was installed into the high-speed wind tunnel of the DLR Göttingen. At different supersonic Mach numbers and at four coolant flow rates in the range of 0–2.5% pressure distribution measurements and probe measurements were performed. The flow field was visualized by schlieren photos and the instantaneous velocity field was quantitatively investigated by Particle Image Velocimetry (PIV). The measurements of the velocity field gave an insight into stationary effects, for example the change of shock strength with coolant flow rate, and instationary effects such as the existence of a vortex street in the wake. The PIV technique offers special advantages for the investigation of transonic flow fields, but also yields to special experimental difficulties, which are also described in this article. Measured losses display a maximum at the downstream Mach number 1. This is strongly related to the behaviour of the base pressure. A loss minimum is achieved at moderate coolant flow rates, showing that an optimum coolant flow rate exists. The loss was analysed and separated into the loss contributions from the profile upstream of the trailing edge and the mixing loss due to the coolant flow.     

Spurious counts in gas volume flow measurements by means of turbine meters

Acoustical oscillations can induce a rotation of a turbine flow meter in the absence of main flow, which leads to spurious counts. A simplified model is presented which explains the occurrence of spurious counts in the limit of very thin turbine blades and high Strouhal numbers. The predicted threshold for the occurrence of spurious counts is compared to experimentally obtained data at various gas pressures in the range from 1 to 8 bar. The simplified model provides a reasonable prediction of the occurrence of spurious counts and can be used as an useful engineering tool in the prediction of the occurrence of spurious counts.     

PIV measurements in a weakly separating and reattaching turbulent boundary layer

A high Reynolds number flat plate turbulent boundary layer was studied in a wind-tunnel experiment using particle image velocimetry (PIV). The flow is subjected to an adverse pressure gradient (APG) which is designed such that the boundary layer separates and reattaches, forming a weak separation bubble. With PIV we are able to get a more complete picture of this complex flow phenomenon. The view of a separation bubble being composed of large scale coherent regions of instantaneous backflow occurring randomly in a three-dimensional manner in space and time is verified by the present PIV measurements. The PIV database was used to test the applicability of various velocity scalings around the separation bubble. We found that the mean velocity profiles in the outer part of the boundary layer, and to some extent also the Reynolds shear-stress, are self-similar when using a velocity scale based on the local pressure gradient. The same can be said for the so called Perry–Schofield scaling, which suggests that the two velocity scales are connected. This can also be interpreted as an experimental evidence of the claimed relation between the latter velocity scale and the maximum Reynolds shear-stress.     

Simultaneous density-field visualization and PIV of a shock-accelerated gas curtain

We describe a highly-detailed experimental characterization of the Richtmyer-Meshkov instability (the impulsively driven Rayleigh-Taylor instability) (Meshkov 1969; Richtmyer 1960). In our experiment, a vertical curtain of heavy gas (SF₆) flows into the test section of an air-filled, horizontal shock tube. The instability evolves after a Mach 1.2 shock passes through the curtain. For visualization, we pre-mix the SF₆ with a small (∼10⁻⁵) volume fraction of sub-micron-sized glycol/water droplets. A horizontal section of the flow is illuminated by a light sheet produced by a combination of a customized, burst-mode Nd:YAG laser and a commercial pulsed laser. Three CCD cameras are employed in visualization. The “dynamic imaging camera” images the entire test section, but does not detect the individual droplets. It produces a sequence of instantaneous images of local droplet concentration, which in the post-shock flow is proportional to density. The gas curtain is convected out of the test section about 1 ms after the shock passes through the curtain. A second camera images the initial conditions with high resolution, since the initial conditions vary from test to test. The third camera, “PIV camera,” has a spatial resolution sufficient to detect the individual droplets in the light sheet. Images from this camera are interrogated using Particle Image Velocimetry (PIV) to recover instantaneous snapshots of the velocity field in a small (19 × 14 mm) field of view. The fidelity of the flow-seeding technique for density-field acquisition and the reliability of the PIV technique are both quantified in this paper. In combination with wide-field density data, PIV measurements give us additional physical insight into the evolution of the Richtmyer-Meshkov instability in a problem which serves as an excellent test case for general transition-to-turbulence studies. Received: 26 June 1999/Accepted: 29 October 1999     

PIV measurements of flow in and around scour holes

Two sets of experiments related to the scour of cohesionless sediment by planar turbulent jets are presented and discussed. The first set of experiments measures the growth of the scour hole and downstream dune as a function of time. Measurements reveal a bedform that is nearly self-similar and whose growth in time is governed by a power-law relationship. The bedform is well represented by three linear segments with slopes near the angle of repose of the sediment. The second set of experiments uses Particle Image Velocimetry to characterize the mean velocity field in the scour hole and above the dune. For this set of experiments, a series of successively larger roughened fixed-bed models was used in place of the mobile bed. The measurements reveal the presence of strong recirculation in the hole and an attached wall jet on the main slope. Discussion of the utility of the present fixed-bed measurements in estimating shear stress along the bed and related application to predictive modeling of hydraulic scour is provided. Discussion of the technical challenges of similar mobile-bed measurements is also given.     

Simultaneous PIV and pulsed shadow technique in slug flow: a solution for optical problems

A recent technique of simultaneous particle image velocimetry (PIV) and pulsed shadow technique (PST) measurements, using only one black and white CCD camera, is successfully applied to the study of slug flow. The experimental facility and the operating principle are described. The technique is applied to study the liquid flow pattern around individual Taylor bubbles rising in an aqueous solution of glycerol with a dynamic viscosity of 113×10^–3 Pa s. With this technique the optical perturbations found in PIV measurements at the bubble interface are completely solved in the nose and in annular liquid film regions as well as in the rear of the bubble for cases in which the bottom is flat. However, for Taylor bubbles with concave oblate bottoms, some optical distortions appear and are discussed. The measurements achieved a spatial resolution of 0.0022 tube diameters. The results reported show high precision and are in agreement with theoretical and experimental published data.Symbols D internal column diameter (m) - g acceleration due to gravity (m s^–2) - l _w wake length (m) - Q _v liquid volumetric flow rate (m³ s^–1) - r radial position (m) - r * radial position of the wake boundary (m) - R internal column radius (m) - U _s Taylor bubble velocity (m s^–1) - u _z axial component of the velocity (m s^–1) - u _r radial component of the velocity (m s^–1) - z distance from the Taylor bubble nose (m) - Z * distance from the Taylor bubble nose for which the annular liquid film stabilizes (m) Dimensionless groups Re Reynolds number ( ) - N _f inverse viscosity number ( ) Greek letters liquid film thickness (m) - liquid kinematic viscosity (m² s^–1) - liquid dynamic viscosity (Pa s) - liquid density (kg m^–3)     

Development of a high-speed UV particle image velocimetry technique and application for measurements in internal combustion engines

A flexible, high-frame rate particle image velocimetry technique that can be applied to operating internal combustion engines in highly luminous combustion situations was developed. Two high-repetition rate diode-pumped Nd:YAG lasers operated at 355 nm and a CMOS camera were used to devise a system that allowed measurements of velocity fields near the spark plug in a firing engine at a rate of 6 kHz for 500 consecutive cycles. The 6 kHz acquisition rate enables recording one velocity field every other crank angle at 2,000 RPM engine speed. Sample results such as individual and average flow fields and kinetic energy evolutions are presented.