Direct comparison of 2D PIV and stereoscopic PIV measurements

Abstract

A stereoscopic particle image velocimetry (PIV) measurement system based onthe translation configuration was developed and applied to the flow behind aforward-swept axial-fan with five blades in a water tank. The 3D calibrationprocedure was employed to compensate the distortion and refraction of particleimages. The perspective error caused by the out-of-plane motion was estimated bydirect comparison of the 2D PIV and stereoscopic PIV (SPIV) results. The SPIVand 2D PIV comparison was carried out for the particle images capturedsimultaneously. The difference of mean velocity data measured by 2D PIV andSPIV techniques is nearly proportional to the mean out-of-plane velocitycomponent. The difference appears to be relatively large in the region nearthe fan blade having higher out-of-plane motion. The perspective errorestimated from the calibration data and mean out-of-plane velocity dataagrees with the mean velocity difference between the 2D PIV and SPIVresults. The turbulence intensity measured by the 2D PIV method isoverestimated in comparison to that of the SPIV method due to the projectedvelocity fluctuations of the out-of-plane velocity component. The turbulentfluctuations caused by the out-of-plane velocity component were estimated bymultiplication of the incident angle obtained in the calibration procedure andout-of-plane velocity fluctuations. The estimated perspective fluctuation ofthe out-of-plane velocity component was compared with the differencemeasured by the two methods. In SPIV measurements of turbulent statistics,therefore, the ratio of out-of-plane to in-plane RMS error determined from thesystem set-up should be considered in order to obtain reliable results.