A mechanistic model that predicts the separation of oil-water dispersed horizontal pipe flows was investigated. Different droplet diameter averages were implemented in the model and the accuracy of the resulting predictions was assessed by comparing each case against experimental data. The experimental data used was obtained in a pilot scale two-phase flow facility using tap water and oil (828 kg m-3, 5.5 mPa s) as test fluids. The results show that the separation length is highly sensitive to the drop diameter, but further investigation is required to determine which drop diameter average produces more accurate predictions of the flow profile.
The study of the separation dynamics of oil/water dispersions is of significant importance to the petroleum industry. Such dispersions are often formed on offshore platforms during pressure reduction and may be stabilised by surfactants found naturally in crude oil. Nevertheless, the density difference between the phases is often enough to cause separation which eventually leads to stratification. This is especially the case during transportation of the extracted crude oil over long distances, for example from the offshore platform to the refinery, and in pipe separators where low velocities exist.