Experimental study of multiphase flow of viscous oil, gas and sand in horizontal pipes

Abstract

The oil and gas industry relies on multiphase flow models and correlations to predict the behavior of fluids through wells and pipelines. Significant amount of research has been performed on the multiphase flow of different types of liquids with gases to extend the applicability of existing models to field-specific fluid conditions. Heavy oil and gas flow research commenced in the past decade and new correlations have been developed that define their flow behavior/regimes. This study aims to plant a foot in the quite deficient area of multiphase flow research that focuses on a sufficiently common situation faced by many heavy oil producing fields: the presence of sand in wells and pipelines. This study will be the first recorded attempt to understand the multiphase flow of heavy oil, gas, and sand. A 1.5" diameter multiphase flow loop facility capable of handling solids was designed and constructed for the study. Data logging instruments were calibrated and installed to be able to withstand the erosive effects of sand. The flow loop was leak and pressure tested with water and air. Three oils of 150, 196 and 218 cP viscosities were utilized to gather 49 single phase liquid, 227 two-phase liquid- air and 87 three-phase liquid, air and solid multiphase flow data points which included differential and absolute pressures, fluid flow rates, temperatures, liquid and composite liquid- solid hold- up data and photo and videotaping of the observed flow regimes. Validation of the setup was performed using single phase flow of oil and two-phase flow of oil and air. Sand was added in three different concentrations to the 218 cP oil and three-phase oil, gas and sand multiphase flow tests were performed. Flow patterns were identified and flow pattern maps were created using acquired data. No change was observed on flow pattern transitions by changing oil viscosities. Liquid hold- up and differential pressures were compared to observe the effect of changing oil viscosity and the presence of sand in varying concentrations on the two-phase flow of oil and gas and the three-phase flow of oil, gas and sand respectively. An increase in differential pressures was observed with increasing viscosities and the addition of sand. No changes in hold-up were seen with changing oil viscosities rather flow patterns impacted liquid hold-up significantly. The slug flow pattern was analyzed. Composite liquid-solid hold-up in slug flow were physically measured and predicted. Liquid slug lengths were predicted and compared with observed lengths using photo and videography techniques. Differential pressures and liquid hold-up were compared with existing multiphase flow models in the PIPESIM multiphase flow simulator to test model predictions against observed flow data. The dependence of differential pressure gradients and liquid hold-up on dimensionless variables was realized by performing normalized linear regressions to identify the most significant dimensionless groups and the results were given a mathematical form by proposing correlations for differential pressure and hold-up predictions. To the best of our knowledge, this study is the first attempt at systematically measuring pressure drop and liquid hold up during the three-phase flow of oil, gas and sand.