An Approach for Judging Confident PIV Measurement in Two-Phase Turbulent Near-Wake Flow

抄録

Particle image velocimetry (PIV) is a non-intrusive optical diagnostic in which the properties of fluid flow are made of the foreign or tracer particles (seedings), and it is, thus, an indirect measurement method. Twophase flow essentially consists of the continuous phase (carrier fluid) and dispersed phases (laden coarse particles or bubbles). The turbulent near-wake flow after a long circular cylinder laden with a binary-size group of particles is employed for demonstrating a two-phase complex flow case in the study. The fine particles (with the mean size of 2.7μm) serve as the seedings and their motion represents the flow behavior of carrier fluid, while the coarse particles (with the mean size of 55 μm) represent the dispersed phase. A double-discriminating process in terms of gray level and size of image pattern together with the median mask technique is employed for discriminating the image patterns of the carrier and dispersed phases in the flow field. However, confident measurements of turbulent flow properties using the PIV diagnostic, no matter what are of the continuous or dispersed phase, must meet the statistical conditions of sufficient numbers of particles falling within each interrogation window of the image. An approach in terms of gray-level ratio of the image is developed to judge attainment of statistically stationary status of the measured two-phase turbulent flow properties, particularly in the very upstream subregion of near wake. The thresholds of gray-level ratio set for the seedings and laden coarse particles are 0.58 and 0.63, respectively, in the study to assure statistically stationary results. It is shown that there indeed exists a core subregion in the very upstream near-wake where PIV diagnostic is incapable of providing confident measurements of turbulent two-phase flow properties.