Geometric scattering removal in cross-correlation Doppler global velocimetry by structured illumination

Abstract

The biases introduced by background light contributions in planar, Doppler-based velocimetry are significantly reduced by the application of structured illumination. The method works by measuring and separating desired flow-scattered signals from unwanted reflections and secondary scattering contributions via a sinusoidal intensity modulation across the laser sheet used to interrogate the flow. In the current work we consider the influence of background light on cross-correlation Doppler global velocimetry (CC-DGV) and present the results for measurements in a turbulent free jet with an exit Mach number of 0.5 in a measurement arrangement with high intensity background contributions. It is shown that if the ratio of the particle-scattering (i.e. Mie scattering) intensity to the background solid surface scattering (i.e. geometric scattering) intensity is greater than 10, then the measured signal displays a significant bias away from the true velocity. With structured illumination, the measured intensity corresponding to geometric scattering is reduced to negligible levels, reminiscent of a measurement arrangement with little or no background light influence. Centreline velocity measurements with this background subtraction method are compared with Kiel-type pressure probe measurements, yielding a root-mean-square (RMS) difference between the two of 1.2 m s -1 (less than 1% of the full scale velocity), well within the published uncertainty bound of the CC-DGV technique, underscoring the robustness of this approach. In addition to the centreline validation, the streamwise vorticity and total pressure are also analysed, showing trends indicative of the S-type shape of the duct and aggressive flow entrainment experienced in this environment.

13th International Symposium on Particle Image Velocimetry (ISPIV 2019)

“…The portfolio of variants of the PIV techniques is continuously growing with the introduction of novel particle-image-based methods and with the maturation of most-recently introduced tools. Boyda et al [12] proposed to apply a structured laser illumination planar imaging, based on applying a sinusoidal intensity modulation along the width of a laser sheet, for removing the background in cross-correlation Doppler global velocimetry, and tested the method on a turbulent jet flowing near a curved wall boundary. Charonko et al [13] proposed to use windowed Fourier transform with local spectrum analysis to analyse molecular tagging velocimetry images and compared its performances in terms of modulation transfer function and error with cross-correlation PIV processing.…”