Bulletin of the American Physical Society
70th Annual Meeting of the APS Division of Fluid Dynamics
Volume 62, Number 14
Sunday–Tuesday, November 19–21, 2017; Denver, Colorado
Session F24: Experimental Methods II: Lasers, Particle Tracking, PIVExperimental
|
Hide Abstracts |
Chair: Leah Mendelson, Harvey Mudd College Room: 703 |
Monday, November 20, 2017 8:00AM - 8:13AM |
F24.00001: A New Approach to Time-Resolved 3D-PTV Aaron Boomsma, Dan Troolin, Dan Bjorkquist Volumetric three-component velocimetry via particle tracking is a powerful alternative to TomoPIV. It has been thoroughly documented that compared to TomoPIV, particle tracking velocimetry (PTV) methods (whether 2D or 3D) better resolve regions of high velocity gradient, identify fewer ghost particles, and are less computationally demanding, which results in shorter processing times. Recently, 3D-PTV has seen renewed interest in the PIV community with the availability of time-resolved data. Of course, advances in hardware are partly to thank for that availability---higher speed cameras, more effective memory management, and higher speed lasers. But in software, algorithms that utilize time resolved data to improve 3D particle reconstruction and particle tracking are also under development and advancing (e.g. shake-the-box, neighbor tracking reconstruction, etc.). .In the current study, we present a new 3D-PTV method that incorporates time-resolved data. We detail the method, its performance in terms of particle identification and reconstruction error and their relation to varying seeding densities, as well as computational performance. [Preview Abstract] |
Monday, November 20, 2017 8:13AM - 8:26AM |
F24.00002: Multi-time lag type PIV analysis of low Reynolds number inhalant flows Aaron True, John Crimaldi Flows with high dynamic ranges of velocity can be challenging to measure with particle image velocimetry (PIV) since there is no single correlation timescale that completely optimizes particle displacements (correlation peak strength) everywhere. Inhalant flows, which draw fluid from a reservoir through an orifice and into a tube under the action of suction, exemplify this high dynamic range condition due to the rapid spatial attenuation of velocity with distance from the inhalant orifice. For these axisymmetric flows with relatively persistent spatiotemporal structure, the primary cause of correlation peak degradation is in-plane loss of particles due to extremely high (low) local displacements. Thus, during PIV analysis of these flows in an index of refraction-matched flow facility (borosilicate glass tubing matched with mineral oil), we developed and applied a simple multi-time lag type PIV postprocessing algorithm which cross-correlates image pairs across two different timescales optimized for some target low and high velocities. This yielded flow fields which were locally correlation-optimized throughout the entire experimental volume. For flows spanning three Reynolds numbers (1, 10, and 100) and three orifice heights (distance to a nearby bottom bed for an upward oriented tube, 0, 2, and 40 inhalant orifice diameters), complimentary numerical simulations showed that our multi-time lag type PIV analysis produced close agreement between measured and simulated flows throughout the experimental volume covering a high velocity dynamic range. [Preview Abstract] |
Monday, November 20, 2017 8:26AM - 8:39AM |
F24.00003: Simultaneous velocity and pressure quantification using pressure-sensitive flow tracers in air Peng Zhang, Sean Peterson, Maurizio Porfiri Particle-based measurement techniques for assessing the velocity field of a fluid have advanced rapidly over the past two decades. Full-field pressure measurement techniques have remained elusive, however. In this work, we aim to demonstrate the possibility of direct simultaneous planar velocity and pressure measurement of a high speed aerodynamic flow by employing novel pressure-sensitive tracer particles for particle image velocimetry (PIV). Specifically, the velocity and pressure variations of an airflow through a converging-diverging channel are studied. Polystyrene microparticles embedded with a pressure-sensitive phosphorescent dye-platinum octaethylporphyrin (PtOEP)-are used as seeding particles. Due to the oxygen quenching effect, the emission lifetime of PtOEP is highly sensitive to the oxygen concentration, that is, the partial pressure of oxygen, in the air. Since the partial pressure of oxygen is linearly proportional to the air pressure, we can determine the air pressure through the phosphorescence emission lifetime of the dye. The velocity field is instead obtained using traditional PIV methods. The particles have a pressure resolution on the order of 1 kPa, which may be improved by optimizing the particle size and dye concentration to suit specific flow scenarios. [Preview Abstract] |
Monday, November 20, 2017 8:39AM - 8:52AM |
F24.00004: Stereo Particle Shadow Velocimetry Jeff Harris, Christine Truong, Michael McPhail, Arnie Fontaine An extension of planar particle shadow velocimetry (PSV) to obtain stereoscopic measurements is presented. PSV is a measurement technique comparable to particle image velocimetry (PIV), the difference being the method of flow field illumination. Pulsed LEDs are used for backlighting in PSV, whereas a laser sheet is used in PIV. PSV is a useful alternative to PIV when there is not optical access for a laser sheet while also providing improved safety and reduced cost. The planar method of PSV has been proven a viable substitute for PIV in 2-component measurements, but the method has not yet been benchmarked for 3-component measurements, such as in stereoscopic imaging. A comparison of stereo PIV and stereo PSV is presented for flow from a simple round jet. The jet was situated in-line with the focal plane (flowing parallel to the laser sheet) and off the focal plane flowing through the laser sheet, thus giving a measurable third component of velocity. [Preview Abstract] |
Monday, November 20, 2017 8:52AM - 9:05AM |
F24.00005: Characterization of turbulent flow structures in the atmospheric boundary layer through super-large-scale particle image velocimetry Michael Heisel, Yun Liu, Teja Dasari, Alec Petersen, Jiarong Hong, Filippo Coletti, Michele Guala Super-large-scale particle image velocimetry (SPIV) using natural snowfall has previously been shown to be a reliable field measurement technique for near-surface atmospheric flows (Toloui et al. \textit{Exp. Fluids}, 55:1737, 2014; Hong et al. \textit{Nature Comm.} 5:4216, 2014). Here we present results from SPIV measurements in the thermally neutral atmospheric surface layer. The data were collected at the EOLOS field station over relatively flat, snow-covered farmland, allowing the development of a fully rough wall boundary layer with a Reynolds number $Re_\tau \sim \mathcal{O}(10^6)$. The data include three time-resolved 15-minute acquisition periods with a field of view extending from 3 m to 19 m above the ground and up to 14 m wide. The flow statistics are validated and supplemented by sonic anemometry from a meteorological tower immediately downstream of the SPIV field of view. The time-resolved planar measurements provide temporal and spatial characterization of key wall turbulence features at high Reynolds number, including ramp-like structures, spanwise vortices, and uniform momentum zones. In comparing the findings to laboratory studies, Reynolds number similarity and the scaling behavior of characteristic properties are discussed. [Preview Abstract] |
Monday, November 20, 2017 9:05AM - 9:18AM |
F24.00006: Non-iterative volumetric particle reconstruction near moving bodies Leah Mendelson, Alexandra Techet When multi-camera 3D PIV experiments are performed around a moving body, the body often obscures visibility of regions of interest in the flow field in a subset of cameras. We evaluate the performance of non-iterative particle reconstruction algorithms used for synthetic aperture PIV (SAPIV) in these partially-occluded regions. We show that when partial occlusions are present, the quality and availability of 3D tracer particle information depends on the number of cameras and reconstruction procedure used. Based on these findings, we introduce an improved non-iterative reconstruction routine for SAPIV around bodies. The reconstruction procedure combines binary masks, already required for reconstruction of the body's 3D visual hull, and a minimum line-of-sight algorithm. This approach accounts for partial occlusions without performing separate processing for each possible subset of cameras. We combine this reconstruction procedure with three-dimensional imaging on both sides of the free surface to reveal multi-fin wake interactions generated by a jumping archer fish. Sufficient particle reconstruction in near-body regions is crucial to resolving the wake structures of upstream fins (i.e., dorsal and anal fins) before and during interactions with the caudal tail. [Preview Abstract] |
Monday, November 20, 2017 9:18AM - 9:31AM |
F24.00007: Key aspects in the implementation of algorithms for digital processors in velocity and size measurements Dan Troolin, Jim Evenstad, Wing Lai Digital signal processing techniques are used to extract accurate flow and size information in complex and difficult measuring situations. Some of the key aspects that are vital to the performance of signal processors are discussed. Limitations of theoretical approach in evaluating the processing techniques are outlined. Simulations have been carried out to examine the influence of some of the parameters not covered by the analytical approach. The robustness of the auto-correlation technique with quadrature mixing is demonstrated through simulations. The advantage of adapting advantage of adapting a processing technique to be well suited to the nature of the signal, as well as the importance of pre-processing or conditioning the input to be properly positioned for the algorithm are pointed out. Finally the benefits of the auto-correlation technique are proven through experimental measurements. [Preview Abstract] |
Monday, November 20, 2017 9:31AM - 9:44AM |
F24.00008: Compact 3D Camera for Shake-the-Box Particle Tracking Christina Hesseling, Dirk Michaelis, Jan Schneiders Time-resolved 3D-particle tracking usually requires the time-consuming optical setup and calibration of 3 to 4 cameras. Here, a compact four-camera housing has been developed. The performance of the system using Shake-the-Box processing (Schanz et al. 2016) is characterized. It is shown that the stereo-base is large enough for sensible 3D velocity measurements. Results from successful experiments in water flows using LED illumination are presented. For large-scale wind tunnel measurements, an even more compact version of the system is mounted on a robotic arm. Once calibrated for a specific measurement volume, the necessity for recalibration is eliminated even when the system moves around. Co-axial illumination is provided through an optical fiber in the middle of the housing, illuminating the full measurement volume from one viewing direction. Helium-filled soap bubbles are used to ensure sufficient particle image intensity. This way, the measurement probe can be moved around complex 3D-objects. By automatic scanning and stitching of recorded particle tracks, the detailed time-averaged flow field of a full volume of cubic meters in size is recorded and processed. Results from an experiment at TU-Delft of the flow field around a cyclist are shown. [Preview Abstract] |
Monday, November 20, 2017 9:44AM - 9:57AM |
F24.00009: Improving accuracy of Plenoptic PIV using two light field cameras Brian Thurow, Timothy Fahringer Plenoptic particle image velocimetry (PIV) has recently emerged as a viable technique for acquiring three-dimensional, three-component velocity field data using a single plenoptic, or light field, camera. The simplified experimental arrangement is advantageous in situations where optical access is limited and/or it is not possible to set-up the four or more cameras typically required in a tomographic PIV experiment. A significant disadvantage of a single camera plenoptic PIV experiment, however, is that the accuracy of the velocity measurement along the optical axis of the camera is significantly worse than in the two lateral directions. In this work, we explore the accuracy of plenoptic PIV when two plenoptic cameras are arranged in a stereo imaging configuration. It is found that the addition of a 2$^{\mathrm{nd}}$ camera improves the accuracy in all three directions and nearly eliminates any differences between them. This improvement is illustrated using both synthetic and real experiments conducted on a vortex ring using both one and two plenoptic cameras. [Preview Abstract] |
Monday, November 20, 2017 9:57AM - 10:10AM |
F24.00010: 3D Rainbow Particle Tracking Velocimetry. Andres A. Aguirre-Pablo, Jinhui Xiong, Ramzi Idoughi, Abdulrahman B. Aljedaani, Xiong Dun, Qiang Fu, Sigurdur T. Thoroddsen, Wolfgang Heidrich A single color camera is used to reconstruct a 3D-3C velocity flow field. The camera is used to record the 2D (X,Y) position and colored scattered light intensity (Z) from white polyethylene tracer particles in a flow. The main advantage of using a color camera is the capability of combining different intensity levels for each color channel to obtain more depth levels [1]. The illumination system consists of an LCD projector placed perpendicularly to the camera. Different intensity colored level gradients are projected onto the particles to encode the depth position (Z) information of each particle, benefiting from the possibility of varying the color profiles and projected frequencies up to 60 Hz. Chromatic aberrations and distortions are estimated and corrected using a 3D laser engraved calibration target. The camera-projector system characterization is presented considering size and depth position of the particles. The use of these components reduces dramatically the cost and complexity of traditional 3D-PTV systems. [1] Xiong, J., Idoughi, R., Aguirre-Pablo, A.A., Aljedaani, A. B., Dun, X., Fu, Q., Thoroddsen, S. T., Heidrich, W. ``Rainbow Particle Imaging Velocimetry for Dense 3D Fluid Velocity Imaging.'' ACM Trans. Graph. 36, 4, Article 36 (2017). [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700