Phase Inversion Correlations Analysis for Oil–Water Flow in Horizontal Pipes

Abstract

Several models and correlations have been developed to predict phase inversion. However, most of them are for specific and limited ranges of operation. Therefore, this research studies 26 phase inversion correlations, five direct phase inversions, and 21 dependent mixture-based correlations using 645 data points from the literature to find the best ones for the prediction of phase inversion. Several flow velocities and fluid properties were taken into account. The results showed that the correlations performed depended on the viscosity range; therefore, for oils between 0.001 and 0.002, the Ngan (2011) model using Einstein’s (1906) or Taylor’s (1936) mixture viscosity correlations was the most reliable correlation. In the case of an oil viscosity between 0.02 and 0.04 Pa s, the direct correlation of Yeh et al. (1964) and the model of Ngan (2011) using Vermeulen et al.’s (1955) mixture viscosity correlation were found to be more accurate.