A Modified Version of the Aziz et al. Multiphase Flow Correlation Improves Pressure Drop Calculations in High-Rate Oil Wells

Abstract

Abstract The prediction of multiphase pressure drop during the simultaneous flow of oil, gas and water in vertical tubing strings is crucial in the development and optimum exploitation of an oil field. In this paper a method is proposed to improve pressure drop predictions by Aziz et al. multiphase vertical-flow correlation. This correlation is theoretically justified as compared to the traditional empirical methods. The present method suggests combining several flow pattern maps with the Aziz et al. correlation in an attempt to achieve the improvement sought. Two field data sets comprising 32 production tests gathered from three different sources in the Middle East and North Africa were used to examine the performance of the various combinations. The results of this work indicate that the performance of Aziz et al. multiphase correlation can be best improved by replacing its original flow-pattern map with the traditional Duns-Ros flow-regime map and for both data sets used. A significant improvement has been observed giving an overall absolute average percent deviation of 2.16% compared with 5.33% for the original correlation. Also, with this combination the relative performance factor has been reduced to 1.33 from 2.90 for the original Aziz et al. correlation. The efficiency of the Aziz et al.-Duns and Ros combination at high oil production rates, when compared with the original Aziz et al. correlation, was further examined by noticing the improvement gained in statistical measures at higher levels of liquid superficial velocity. This work represents an addition to the technology of multiphase behavior in vertical pipes and its results will help in a more accurate design of tubing strings in high-rate producing oil wells.