Micro-PIV (μPIV) has the potential to be applied to combusting turbulent poly-disperse sprays, and can be considered as an extension to a standard shadow sizing technique. Commercial software which allows for the determination of both the size and velocity of the observed fragments is readily available. However, a key limitation of these shadow based techniques is the dependence on the measurement depth of the lens due to the absence of a planar light sheet. Whilst this presents a limitation, it is also the key advantage of μPIV, given the removal of droplet scattering and the ability to instantaneously visualize both the size and velocity of arbitrarily shaped fragments. In this contribution, by studying sprays dominated by spherical objects and comparing with LDA/PDA, we examine the limitations of applying μPIV towards the characterization of velocities in dilute poly-disperse turbulent flows with different initial velocity profiles. The results indicate that the μPIV technique can be biased towards larger droplet sizes, providing an effective ‘size sub-ranged’ velocity profile which is linked to the limited spatial resolution of the lens. By binarizing the images with a threshold closer to the background, the bias is reduced in the centreline, but leads to a reduction of the out-of-plane resolution due to an increase in the measurement depth. These results suggest the need for a balanced normalized background threshold for accurate resolution of planar velocities. Velocity profile shapes can be accurately recovered, and, with an accurate threshold choice, the magnitudes of the turbulence intensity and centerline mean velocities are in excellent agreement with PDA results. The measurements provide insights into the limitations of using μPIV in poly-disperse turbulent sprays and guide future development in this area.
|Number of pages||4|
|Journal||ACS 2015 : Australian Combustion Symposium : proceedings|
|Publication status||Published - 2015|
|Event||Australian Combustion Symposium - Melbourne|
Duration: 7 Dec 2015 → 9 Dec 2015
- Image Processing