Evaluation of Models for Droplet Shear Effect of Centrifugal Pump

During the process of petroleum production and transportation, equipment such as pumps and chokes will cause shear effects which break the dispersed droplets into smaller size. The smaller droplets will influence the separator process significantly and the droplet size distribution has become a critical criterion for separator design. In order to have a better understanding of the separation efficiency, estimation of the dispersed-phase droplet size distribution is very important. The objective of this paper is to qualitatively and quantitatively investigate the effect of shear imparted on oil-water flow by centrifugal pump. This paper presents available published models for the calculation of droplet size distribution caused by different production equipment. Also detailed experimental data for droplet size distribution downstream of a centrifugal pump are presented. Rosin-Rammler and Log-Normal Distributions utilizing dmax Pereyra (2011) model as well as dmin Kouba (2003) model are used in order to evaluate the best fit distribution function to simulate the cumulative droplet size distribution. The results confirm that applying dmax Pereyra (2011) model leads to Rosin-Rammler distribution is much closer to the experimental data for low shear conditions, while the Log-Normal distribution shows better performance for higher shear rates. Furthermore, the predictions of Modified Kouba (2003) dmin model show good results for predicting the droplet distribution in centrifugal pump, and even better predictions under various ranges of experiments are achieved with manipulating cumulative percentage at minimum droplet diameter F(Dmin).