Abstract:
Multiphase flow in the oil and gas industry covers a wide range of flows. Thus, over the
last decade, the investigation, development and use of multiphase flow metering system
have been a major focus for the industry worldwide. However, these meters do not
perform well in slug flow conditions.
The present work involves experimental investigations of multiphase flow measurement
under slug flow conditions. A two-phase gas/liquid facility was designed and
constructed at Cranfield University. It consisted of a 0.05 m diameter 25 m long
horizontal pipeline with the necessary instrumentation.
An ultrasonic multiphase metering concept has been proposed and investigated. The
concept was based on the combination of non-invasive and non-intrusive ultrasonic
sensors and a slug closure model. The slug closure model was based on the "slug unit"
model to infer the gas and liquid phase volumetric flowrates.
The slug characteristics obtained by non-invasive and non-intrusive ultrasonic
techniques were inputs to slug closure model which calculates the factors KI
(Liquid),
K2
(Liquid),
K3
(Gas) and K4
(Gas).
These factors are function of the slip ratio in the slug body,
flow profile (CO), drift velocity (Vd), liquid holdup and gas void fraction in slug body,
slug length, film length, and the total length of the slug unit. Based on ultrasonic sensor
measurements, the slug translational velocity was estimated and the slug closure model
then calculates the gas and liquid phase volumetric flowrates.
Air water slug flow data were gathered and processed for a range of superficial
velocities VSL=0.3 to 1.03 ms'1 and VsG=0.6 to 3.01 ms'1. The overall goal of a 5%
relative error metering for both phases was not achieved for the conditions tested. The
liquid phase percentage errors were from
-63.6% to 45.4% while the gas phase
percentage errors were from 42% to -14.6%.
Key words: slug flow, slug characteristics, slug closure model, non-invasive ultrasonic,
non-intrusive ultrasonic, clamp-on transit time ultrasonic flowmeter.
