Prediction of Slug Length for High Pressure Gas/Liquid Two-Phase Flow in Horizontal and Slightly Inclined Pipes

Abstract

Slug flow in pipelines is the most common flow pattern. Slug length is crucial characteristic for pipeline and downstream separation facility design and operation. In addition, mechanistic two- phase flow models require slug length as closure relationship to solve for pressure gradient and average liquid holdup in slug flow. However, the existing slug length closure relationships developed for low pressure are found to poorly perform in high-pressure conditions, i.e. high gas- to-liquid density ratio high, resulting in high uncertainty predictions of slug length, pressure gradient and liquid holdup. This work aims to propose a mechanistic slug length model and to identify the optimal closure relationship for high-pressure condition through error minimization technique using Genetic Algorithm. In addition, the identified set of closure relationships are found to match the physics of slug flow under the investigated conditions. As a result, the proposed model result in a coefficient of determination R2 = 0.85 and an Absolute Average Error (AAE) approximately equals 70% outperforming the best-performing exiting model in the literature.