EXPERIMENTAL INVESTIGATIONS AND COMPUTATIONAL FLUID DYNAMIC PREDICTIONS OF PRESSURE DROP CHARACTERISTICS IN MULTIPHASE FLOW FOR CONCENTRIC INCLINED ANNULI IN NEAR HORIZONTAL WELLS

Abstract

In oil and gas sectors, the accuracy of pressure drop predictions plays significant role in designing multiphase flow systems. In recent decades, there is a considerable amount of theoretical and experimental studies dealing with pressure drop predictions for multiphase flow using the hydraulic diameter concept for non-circular cross-section pipes and annular geometries. In single-phase flow applications, applying the hydraulic diameter has been proven to be accurate enough. However, the accuracy of using the hydraulic diameter in multiphase flow systems for annular geometries is questionable and poorly describes the characteristics of the annular flow, which is a research gap that needs to be addressed properly. Recently, the computational modelling has become a significant tool that has proven its reliability in solving and analyzing the dynamics of the fluids and complex multiphase problems. Therefore, in this thesis multiple single- and two-phase flow experiments were conducted at Texas A&M University at Qatar to study the pressure characteristics of liquid-air phase flow in a concentric annulus under different inclination angles to evaluate and judge the theoretical correlations. Then, a computational fluid dynamic model has been developed to simulate a coupled transient two-phase flow using ANSYS software. Two different combinations of liquid-air phase have been experimented using two different liquids (water and Flowzan), the fluids are flowing in an annular pipe with inner and outer diameters of 6.35 cm, and 11.43 cm, respectively, liquid flow rate range from 192 to 315 kg/min and gas injection pressure range from 1 to 2 bar. Repeatability test showed that the closeness of agreement between the results of successive measurements is as high as ≅98.64%. The implementation of the hydraulic diameter concept in single phase flow showed satisfactory results with a maximum error of 3.9%. However, in two-phase flow, a remarkable difference of about 30% to 81% was observed between the actual and predicted pressure drop values using theoretical two-phase flow correlations. Compared with other theoretical approaches, Beggs and Brill correlation performs the best by predicting the results in the error range of ±40%. ANSYS 2021 was used to develop a two-phase computational fluid dynamics (CFD) model using Eulerian-Eulerian as the multiphase flow model and shear stress transport (SST) as the turbulence model. The CFD results were validated using the conducted in this thesis experimental results. The developed CFD model successfully predicted the experimental pressure drop data with an error in the range of ±10% with an absolute average relative error of 8.4%. The experimental and CFD results and approach developed in this thesis serves as a reference for future research to accurately predict the pressure drop for two-phase flow in annular and other complex geometries.