Interaction between Flocculation and Turbulence – a Method Using the Combination of Image Processing and Piv

In this paper a developed method has been introduced to study flocs or agglomerates and their interaction with turbulence. The images have been taken from a flow having particles, which have a tendency to form flocs. PIV (Particle Image Velocimetry) has been used to locate the specific „high turbulence areas“ in the flow. The same images have been processed digitally in order to find flocs or agglomerates. Two different image processing methods for the floc analysis have been tested: (a) Segmentation method (b) Treshold method. The segmentation method is a variation of the socalled region growing method and the treshold method is a variation of the RLC (Run-Length Coding). Two different methods for the definitions of the turbulence have been used: (1) kinetic energy of the turbulence and (2) especially for this purpose generated function in which an instantaneous turbulence field is defined as a product of the acceleration and the angle change of the neighbouring PIVvelocity vectors. When the instantaneous turbulence fields have been time averaged, the high values seem to form an area with most of the vorticies and large-scale eddies of the flow. Finally, it has been experimentally studied how the flocs are located after a sudden expansion in a channel – i.e. if the high turbulence affects the location and the size of the flocs. The specific high turbulence area defined by the method 2 seems to correlate not only with the sizes and locations of the flocs but also with the cross-directional consistency profile of the flow. According to the experiment, the results depend not only on the turbulence but also on the quality and the quantity of the particles in the flow. For this reason, the method presented in this paper is general and experiments shown examples only.