Summary
A comprehensive numerical modeling study was performed to investigate impact
of pattern confinement on steamflood simulation results, using a three-phase
and 3D thermal reservoir simulator. In addition, the effects of cyclic steaming
of the producers, grid size, and other physical parameters were evaluated.
Detailed multipattern, single-sand steamflood models were constructed using
properties of a heavy-oil field in California. All models included an initial
primary depletion zone of 6 ft within 60 ft of net pay. Up to twenty-five,
2.5-acre patterns were included in the study.
Results show that finely gridded models accurately capture near-vertical
steam override and oil drainage by gravity with a near-horizontal steam/oil
interface. High injection pressures observed in many prior simulations are
primarily a result of confined reservoir models. Steam-zone pressures and
temperatures are similar to those typically observed in the field, when the
model is unconfined (i.e., the model area is greater than the pattern area),
representing undeveloped portions of the field. Moreover, including cyclic
steaming of producers accelerates the steam breakthrough time and lowers
injection pressures. During the post-breakthrough steam-rate-reduction period,
field-observed oil-production response is represented better when the influence
of surrounding patterns is included. Production-rate decline is relatively
small when injection rate is reduced only in the primary pattern(s); however,
the decline rate increases if rate reduction is implemented in the entire
field.
Introduction
Reservoir simulation is a tool used by engineers to design new field
projects and to help manage existing ones. It is commonly used to evaluate or
screen various operating strategies. When used properly, it is today’s most
detailed and sophisticated tool in the oil and gas industry. However, as Coats
(1969) emphasized, the level of sophistication and complexity included in
simulation models must be consistent with the overall project objectives and
reliability of the available input data.
The use of simulation in steamflooding projects has been somewhat limited in
the past, mainly because of excessive computation-time requirements for
modeling project areas large enough to represent actual field operations.
Thermal simulations are typically more CPU-intensive in nature than typical
black-oil simulations as they require the solution of energy-balance equations
in addition to mass-balance equations. In the past, computing hardware
capabilities were not adequate to run sufficiently large thermal models.
Consequently, early thermal simulation studies had to be simple, and they
included only small pattern-element models, which could not capture some actual
field observations (Chu and Trimble 1975; Gomaa et al. 1977; Chu 1979, 1987;
Ziegler 1987; Hong 1994). However, because of recent developments in computer
hardware technology, faster CPUs became available. Use of faster CPUs combined
with better numerical solvers, and advancements in parallel computing, resulted
in significant improvements in the performance of thermal simulators, therefore
enabling modeling of larger, multipattern project areas (Dehghani et al. 1995;
Johnson et al. 1992; Kumar 1992; Williams et al. 2001).
© 2007. Society of Petroleum Engineers
View full textPDF
(
1,729 KB
)
History
- Original manuscript received:
16 June 2004
- Meeting paper published:
26 September 2004
- Revised manuscript received:
15 March 2007
- Manuscript approved:
10 May 2007
- Version of record:
20 October 2007
View Cart
