Instantaneous whole field measurement of velocity and size of air microbubbles in two-phase flows using DDPIV

Abstract

Listen

Translate article

Defocusing digital particle image velocimetry (DDPIV) is the natural extension of digital particle image velocimetry (DPIV), planar or quasi three-dimensional, to a true and unique three-dimensional PIV technique. This work presents the defocusing optical concept by which the depth information can be retrieved, thus overriding the limitation to in-plane measurements of actual PIV techniques, either standard or stereo-based. The concept is implemented into a three-dimensional imaging system specifically designed for the purpose of mapping two-phase bubbly flows. Digital images of the bubble field are recorded and analysed to provide information both on the physical location of every single particle/bubble and on its respective size, which is estimated from the scattered light intensity. The calculation of the true three-component velocity field is done by local spatial cross-correlation between two consecutive sets of particle/bubble locations. The spatial resolution and uncertainty limits are established based on a simplified model of the defocusing optical system. Accuracy measurements show that the average error on the displacements is about 0.02 pixels. The methodology used to measure the size is laid out by application of the Mie scattering theory. A DDPIV prototype instrument was fabricated on specific requirements. The instrument records high resolution images of the bubble field and is capable of providing bubble size and bubble location within a cubic foot volume. The technique is applied to the study of the dynamics of sub-millimeter air bubbles in a three-dimensional vortical flow generated by a propeller. Velocity, bubble size distribution and void fraction for these flows are discussed.KeywordsVoid FractionBubble SizeReference PlaneBubbly FlowParticle Tracking VelocimetryThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.