Summary
This paper explores multiple completion options in gas/condensate reservoirs
with compositional simulations. Besides intelligent-well completion (IWC),
options included commingling two reservoirs of contrasting conductivity
(permeability-thickness product) and selectively perforating zones or
reservoirs to offset the permeability contrast. At the outset, a
value-of-information exercise suggested probing downhole sensing and completion
issues in a stacked-reservoir situation. The ultimate objective of this study
was to ascertain economic completion strategy so that depletion of reservoirs
occurs evenly at the project’s termination.
Single-well compositional simulations formed the backbone for our evaluation
of three completion options. Each reservoir was characterized by history
matching drillstem tests (DSTs). Experimental design (ED) reduced the large
number of simulation runs to a manageable few for probabilistic forecasting.
Comparison of three options suggested that all of them nearly produced the
desired results of maximum liquid recovery despite a 10-fold difference in
permeability between the two horizons.
Results further showed that condensate banking was a nonissue in this
high-kh system of reservoirs as far as the gas deliverability is concerned. In
other words, although 40 to 60% degradation in the gas productivity index (PI)
occurred, gas deliverability remained intact. In contrast, both the liquid PI
and rate declined with time owing to phase-behavior and relative permeability
issues. Finally, we learned that the net income generated by IWC is no better
than the specific-perforation completion (SPC).
Introduction
IWC is primarily about proactive, on-time intervention through monitoring
and control of flow in and out of the well. Economic imperatives, particularly
in deepwater settings, have generated intense interest in IWC technology. Of
course, the ability to commingle marginal reservoirs in any situation is
another attractive application of IWC.
To underscore IWC’s importance, various papers have appeared in the
literature describing instrumentation and control (Robinson and Mathieson 1998;
Rundgren et al. 2001; Tourillon et al. 2001), reservoir modeling (Ostvik et al.
2001; Borch 2001; Yu et al. 2000; Akram et al. 2001; Nielsen et al. 2001;
Jalali 1998), and field implementations (Lau et al. 2001; Erlandsen 2000;
Glandt 2003). Most papers describe the well-centric benefits of IWC in
horizontal and/or multilateral wells in complex lithologies, with the exception
of Jalali (1998) and Glandt (2003). In fact, Glandt provides a comprehensive
review of this technology, particularly from the viewpoint of reservoir
engineering. The use of probabilistic analysis (van der Poela and Jansen 2004)
to assess the value of IWC in a complex reservoir scenario was also explored
for oil wells. This body of work demonstrates the promise of IWC because the
technology is in growth mode, with 5 or so years of collective industry
experience.
The main motivation of this study stemmed from understanding the local
regulatory body’s completion philosophy. Derived primarily from the oilwell
analog, the current regulation prevents layer commingling when production
occurs from different geologic horizons. This regulation does not preclude
commingling layers of contrasting properties, separated by shales, so long as
they are within the same geologic unit. Therefore, our incentive was to learn
how one should approach the completion issue within a single geologic unit and
in multiple geologic units.
In this study, we probe the benefits of IWC or its analog in a
depletion-drive system, where primary production dominates. Three completion
scenarios were considered: (1) commingling with downhole control, (2)
commingling without downhole control by selective interval perforating, and (3)
conventional commingled completion. Our objective was to eliminate reservoir
crossflow without differential depletion among the reservoirs in the first two
completion options.
© 2006. Society of Petroleum Engineers
View full textPDF
(
1,811 KB
)
History
- Original manuscript received:
17 December 2004
- Meeting paper published:
26 September 2004
- Revised manuscript received:
2 November 2005
- Manuscript approved:
2 December 2005
- Version of record:
20 February 2006
View Cart
