ID 97

Analysis of droplet clustering in air-assist sprays using Voronoi diagrams

 

Abstract:

The clustering of droplets in polydispersed sprays generated from a twin fluid air-assist atomizer is experimentally studied. The consequence of droplet clustering in sprays can be crucial for wide range of applications ranging from combustion engines to spray drying. The droplet clusters are identified based on application of Voronoi analysis on the spray droplet images captured using planar laser sheet imaging. Measurements are reported at an axial location 30 cm downstream of the injector exit and for different radial stations. A sensitivity analysis is carried out to infer the influence of intensity thresholding (to identify the droplets) on the measured Voronoi cell distribution. Four different liquid mass loading ratios (ϕm) are considered by varying the air and water flowrates. The ILIDS technique is used to measure droplet size distribution. The degree of clustering is estimated by comparing the standard deviation of the normalized Voronoi area with that of a Random Poisson process, and its variation with respect to ϕm clearly indicated the influence of liquid loading on droplet clustering. The droplet clustering tendency was found to increases towards the edge of the spray. The PDF of normalized cluster area suggest that the clustering of droplets occurs over a wide range of turbulent length scales. Also, the PDF shows a decay with a slope of -5/3 for all ϕm and radial measurement stations. The paper also discusses the geometric characterization of clusters, and the relative droplet number density in a cluster and their significance under different liquid mass loading conditions.