A novel approach for conditional measurement of droplet size distribution within droplet clusters in sprays

A novel method for conditional measurement of droplet size distribution within droplet clusters in sprays is reported in this paper. The salient feature of this approach is the application of Voronoi analysis to droplets measured by the ILIDS technique, which is capable of providing planar measurement of size of individual droplets in a spray. The principle of the present technique is based on calculation of Voronoi tessellations around the droplets identified in the ILIDS image processing. Comparison of the distribution of areas of the Voronoi cells with that from a random distribution of droplets described by a Poisson function facilitates classification of the Voronoi cells as clusters or voids. Thus, individual clusters of droplets and void areas are recognized by identifying the interconnected cells. Since the sizes of individual droplets are known, conditional measurement of droplet size distribution within clusters and voids is possible. Since droplet images are always defocused in the ILIDS technique, correct identification of droplet centers is important, which is described in the paper. It is shown that in spite of the low validation ratio in the ILIDS image processing, the results on the cluster and void characterization agree well with those by Voronoi analysis of PIV images. The novel method is demonstrated for an air-assist spray where ILIDS measurements are obtained far downstream of the injector exit. Interestingly, the characteristic droplet size within droplet clusters is found to be smaller than that in voids, indicating size-based preferential segregation of droplets due to clustering in sprays. Graphical abstract