Summary
Transient multiphase flow in the wellbore causes problems in well-test
interpretation when the well is shut in at surface and the pressure is measured
downhole. Pressure-buildup data recorded during a test can be dominated by
transient wellbore effects (i.e., phase change, flow reversal, and re-entry of
the denser phase into the producing zone), making it difficult to distinguish
between true reservoir features and transient wellbore artifacts (Gringarten et
al. 2000).
This paper is a follow-up to paper SPE 96587 (Ali et al. 2005), which
presented experimental results of phase redistribution effects on
pressure-buildup data. Though the results of the experiments were revealing,
they are complex because they reflect the real well situation. To obtain
results in which the phase redistribution in the well is studied independently
of the interaction with the reservoir, a further set of experiments was carried
out. In these experiments, the tube (simulating the well) was isolated at both
the top and the bottom at the same time. The pressure distribution was measured
during the transient following shut-in and for the steady-state final
condition, in which there was a liquid-filled zone at the bottom of the test
section and a gas-filled zone at the top. A substantial number of tests were
conducted in the bubbly-flow region and could therefore be analyzed by a simple
1D model for bubbly flow. The results of the comparison between the model and
the experimental data are presented in this paper.
Introduction
In the first study (Ali et al. 2005), experiments were carried out to
investigate the effects of wellbore phase redistribution (WPR) and phase
re-injection on pressure-buildup data. Single-phase- and two-phase-flow tests
were conducted with air and water in the long-tube system (LOTUS) at Imperial
College. The LOTUS test layout, as described in paper SPE 96587 (Alii et al.
2005), was designed to simulate a reservoir connected, by a resistance, to the
base of a flowing well. The "reservoir" was recreated by a pressurized
vessel, while the "well" was simulated by a 10.8-m -long,
32-mm-internal-diameter vertical pipe (i.e., the main LOTUS test section). The
well was flowed at controlled rates to mimic those encountered in
gas/condensate reservoirs. After steady-state conditions had been attained, the
well was shut in at the top of the rig (i.e., at the surface) and the
associated transient phenomena were monitored through distributed measurements
of pressure, temperature, liquid holdup, and wall shear stress.
Pressure-buildup data were interpreted with established well-test-analysis
techniques.
These initial experiments provided a qualitative and quantitative
understanding of the effects of gas rates, liquid rates, and rising gas bubbles
on WPR and phase re-injection. Gas flow rate was found to have a higher effect
than water flow rate on WPR. This was most probably because of annular flow
being the predominant flow regime for the experiments. Phase re-injection was
simulated successfully. The lower the reservoir pressure, the higher the liquid
re-injection, an analog to low-permeability reservoirs. For a closed system,
WPR took place. Rising gas caused an increase in bottomhole pressure.
The focus of the second study, presented here, was to investigate WPR
independently of the interactions with the reservoir. The LOTUS tube was
isolated at both the top and the bottom at the same time. The test section was
again the LOTUS 10.8-m-long, 32-mm-internal-diameter vertical tube. A two-phase
flow was set up with known air- and water-flow rates. The pressure distribution
and void fraction were measured for the steady-state flow, and the flow
subsequently was shut down by closing valves at the top and bottom of the test
section simultaneously.
Although the experiments covered a wide range of conditions, a substantial
number of tests were conducted in the bubbly-flow regime. A simple, 1D model
for bubbly flow was developed and implemented for comparison with the
experimental data.
Earlier efforts toward understanding the physics of gas-bubble migration in
wells were carried out by Hasan and Kabir (1994; 1993) and Xiao et al. (1996).
Aremu (2005) provided an overview of bubbly-flow models applied to the problem
of gas kicks while drilling. A detailed review of previously published work on
research into transient wellbore phenomena is presented by Falcone (2006). In
recent years, much work has been carried out on the phenomena occurring in
bubbly flows with a wide range of local measurements, and increasingly, many
use computational methods to represent the detailed motions and interfacial
deformations of the bubbles.
© 2008. Society of Petroleum Engineers
View full textPDF
(
568 KB
)
History
- Original manuscript received:
28 June 2006
- Meeting paper published:
24 September 2006
- Revised manuscript received:
26 November 2007
- Manuscript approved:
27 November 2007
- Version of record:
20 September 2008
View Cart
