Pressure drop prediction for stratified oil-water flow in horizontal pipe

Immiscible liquid-liquid flow of oil-water is common phenomenon in many industrial processes; amongst them is crude oil transportation in oil and gas industry. Produced water with oil has complex interfacial structure which complicates the hydrodynamic predictions of the fluid flow. Due to the compl...

Full description

Bibliographic Details
Main Author: Mohd. Alias, Adib Zulhilmi
Format: Thesis
Language:English
Published: 2015
Subjects:
Online Access:http://eprints.utm.my/48802/25/AdibZulhilmiMohdAliasMFM2015.pdf
Description
Summary:Immiscible liquid-liquid flow of oil-water is common phenomenon in many industrial processes; amongst them is crude oil transportation in oil and gas industry. Produced water with oil has complex interfacial structure which complicates the hydrodynamic predictions of the fluid flow. Due to the complexity of liquid-liquid flow, development of reliable analysis tool is difficult. Computational Fluid Dynamics (CFD) has been an established tool for flow analysis in the field of single phase flow but has only started becomes established in multiphase field. Therefore, this thesis attempts to model stratified oil-water flow in horizontal pipe using ANSYS software Fluent. Since pressure drop is an important consideration in liquid-liquid flow, the modelled flow is then used to predict pressure drop base on the factor of superficial velocity at each liquid phase. The results were compared against the established experimental data available in the literature for reliableness. Base on the simulation, it was evident that the Volume of Fluid (VOF) modelling approach is able to model stratified oil-water flow and possible to predict for pressure drop. Generally the predicted pressure drop was in quite good consistency with experimental data for low superficial velocities but predicted low agreement for higher superficial velocities. However, the simulation can be improved as the turbulence model adopted for this simulation could be modified to obtain better pressure predictions.