Impact of inlet slug flow on horizontal gas-liquid separator
This study has demonstrated that the current sizing criteria for a gas-liquid separator commonly used in the industry maybe inadequate. The sizing criteria has not taken into account the inlet flow pattern and fluid hydrodynamics inside the separator. In this study, the inlet flow pattern was sho...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English |
Published: |
2019
|
Subjects: | |
Online Access: | http://psasir.upm.edu.my/id/eprint/83138/1/FK%202019%2067%20-%20ir.pdf http://psasir.upm.edu.my/id/eprint/83138/ |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | This study has demonstrated that the current sizing criteria for a gas-liquid separator
commonly used in the industry maybe inadequate. The sizing criteria has not taken
into account the inlet flow pattern and fluid hydrodynamics inside the separator. In
this study, the inlet flow pattern was showed to have significant impact in the
separator sizing and the fluid hydrodynamics inside the separator. The experiment
was conducted using a test rig consists of a horizontal gas-liquid separator operating
at 45 °C. The liquid was constantly fed at 430 mL/min while the gas flowrate was
adjusted to meet the requirement of inlet momentum from 200 to 2500 Pa. The liquid
and gas phase used in the experiment were water and air, respectively. Experimental
results demonstrated that the inlet flow was slug flow for inlet momentum of 200 to
1500 Pa, and the Mandhane’s map could be used to predict the flow pattern Computational Fluids Dynamics (CFD) has demonstrated its function as an essential
tools to simulate hydrodynamic inside the separator. However, the standard approach
for Volume of Fluids (VOF) does not able to model the hydraulic jump in the inlet
slug flow and additional User Define Function (UDF) using normal distribution
equation was used at the inlet boundary to simulate the hydraulic jump. Another
advantage was demonstrated using CFD in this study is the complex calculation such
as kinetic energy of turbulence can be solved easily. Realizable k-ε turbulence
model has provided good agreement with experimental result and is recommended
for this type of application. At inlet momentum of 800 Pa and lower, the sloshing does not occurred in the
separator and the Souders-Brown constant from the guidelines is still valid for gas
phase section sizing for horizontal gas-liquid separator which provided acceptable low level liquid carryover at gas phase. For inlet momentum of 1000 Pa and higher,
the impact of inlet slug flow initiated the separator sloshing phenomena. Sloshing
was found to increase the liquid carryover in the gas phase. Kinetic energy of
turbulence at liquid phase shall be kept below 15 m2/s2 to avoid any sloshing in the
separator. The sloshing phenomena in the separator can be avoided by lowering the
liquid level in the separator which is a very economical approach in retrofitting gasliquid
separator and to avoid expensive separation internals. Additional
recommended liquid level from this study was suggested in the gas-liquid separator
sizing in order to mitigate the sloshing phenomena in the separator. |
---|