NUMERICAL SIMULATION OF TWO-PHASE SLUG FLOW REGIME IN HORIZONTAL PIPE

• Main Author: SAM, BAN
• Format: Thesis
• Language: English
• Published: 2017
• Subjects: TJ Mechanical engineering and machinery
• Online Access: http://utpedia.utp.edu.my/22082/1/BanSam_MSc_e-Thesis.pdf; http://utpedia.utp.edu.my/22082/

Oil and gas offshore production headers frequently experience random trips which is suspected to be due to liquid carryover issues as a result of sudden slug generation in the production headers. These slug generations are caused by sudden increase in gas velocity which provides sufficient drag to allow the free surface wave to grow on gasliquid interface to a height sufficient to bridge the pipe. This hydrodynamic slugging is difficult to predict since it can occur over a wide range of gas-liquid flow conditions. Compounding the complexity is the issue of no standard consensus of applicability of numerous operational correlations associated with slugging, e.g. pressure gradient, liquid holdup and slugging frequency. The objective of this study is to investigate oilgas slug formation in horizontal straight pipe and its associated pressure gradient, slug liquid holdup and slug frequency. The abrupt change in gas/liquid velocities, which causes transition of flow patterns was analyzed by using FLUENT 16.1. The incompressible Volume of Fluid (VOF) method was used to capture the transient distribution of segregated gas-liquid interface in different flow patterns. The validity of present model and its methodology was compared using Baker chart for 3.15 inches diameter horizontal pipe. The numerical solution was also compared quantitatively with existing experimental data to ensure its correctness. Finally, analyses were carried out for oil and gas vapor flow, covering substantial region of slug flow regime map. Concurrently, slug pressure gradient, liquid holdup and frequency were extracted numerically, and compared to existing empirical correlations. The present results compares favorably with some correlations and identified some outlier empirical models, which may not be suitable for operational usage.