This paper presents successful applications of gas lift technology to
heavy-oil reservoirs in Intercampo oilfield, Lake Maracaibo, Venezuela. Liquid
production rates range from 10 to 320m3/day per well. Gas lift was
selected as the first artificial lift method in the oilfield. The paper
describes the gas lift mechanisms applied in a high-water-cut heavy-oil (below
15 API) reservoir. The theoretical analysis showed that the injection gas rate
for gas lift and the gas/oil ratio (GOR) of an oil well have direct effects on
the fluid flow from the wellbore.
Theoretical design and actual gas lift production are described in the
paper. The correlations used for artificial gas lift design for high-water-cut
heavy oil need to be refined to match the field data. The difference between
theoretical design and actual production is significant for high-water-cut
heavy oil lower than 15 API. Formation of oil/water emulsion was not observed
during gas lifting of low-API, high-water-cut oil from wells.
In this study, a correction coefficient for gas lift design was applied to a
high-water-cut, low-API field. Further work is needed to refine this gas lift
design software. It should prove particularly useful for production engineers
in optimizing the design of gas lifting equipment.
Continuous gas lift has been employed in lifting heavy crude for many years
(Blann et al. 1980; Redden et al. 1974; Boberg et al. 1973; Abdel et al. 1996).
The gas lift method has been widely applied in the former Soviet Union and in
Venezuela (Ferrer and Maggiolo 1991; Apyev 1978; Ametov et al. 1985). In fact,
heavy oil with a density of between 0.934~0.9659g/cm3 and viscosity
lower than 50cp is commonly processed by continuous gas lift in Venezuela.
Experimental investigation shows that when a 3% hydrocarbon solvent is
injected during gas lifting, daily oil production will increase. Actual data
from former Soviet Union oilfields show that if the water cut is lower than
40%, the solvent does the work. If the water cut is higher than 50%, the
hydrocarbon solvent effect is minimal. Solvent will have no effect at all when
the water cut is higher than 70% (Apyev 1978; Ametov et al. 1985).
Generally, rod pumps have proven to be the best artificial lift method in
heavy oil reservoirs, especially for heavy crude oil at densities between 0.96
Some researchers have thought that gas lift is not suitable for lifting
low-API crude (Berevkiy and Pershchev 1982; Diaz 1981; Brown 1982; Clegg et al.
1993);others have said that it is not feasible for lifting low GOR oil because
the gas rate from the field may be too low to support the gas lift operation
(Douglas et al. 1989; Johnson 1968). However, gas lift has been successfully
applied to this type of well in Venezuela with good results. It has been
suggested that gas lift, if applied appropriately, could be the best artificial
lift method for heavy oil with water cut from an economic point of view.
The literature indicates that the fluid-flow behavior in gas lift resembles
the natural flow in a vertical or near-vertical well. (“Executive Committee”
1984; Palke 1996; Begges 1991; Pengju 2003; Baker-Hughes 2003). However, as a
matter of fact, change of phase regimes induced by gas lift are much more
complex than those in natural flow, because high-velocity gas, as it enters the
well tubing through the gas lift valve and then mixes with oil, gas, and water
in the reservoir fluid, brings in not only an external gas mass but also an
external energy supplement. The high-velocity flow creates a new multiphase
fluid regime, changing from the liquid phase to a continuous gas phase
(transitional flow). Gas bubbles join together and liquid may be entrained into
the bubbles. Although the liquid-phase effects are significant, the gas-phase
effects are dominant. In a later stage, annular flow, mist flow, or both occur,
the gas phase becomes continuous, and the liquid is entrained as droplets in
the gas phase. Gas phase controls the pressure gradient rather than following
the three types of flow regime in a gas/liquid flow typical of vertical tubing.
The resulting multiphase flow consists of bubble, slug, plug, and mist
flow all the way from the bottomhole to the wellhead. Factors influencing the
flow regime include borehole deviation, proportion of each phase, relative
differences in phase densities, surface tension and viscosity of each phase,
average velocity, tubing roughness, and chock size. Kickoff pressure and
casinghead pressure are functions of the previously mentioned factors.
© 2007. Society of Petroleum Engineers
View full textPDF
- Original manuscript received:
15 July 2005
- Meeting paper published:
1 November 2005
- Revised manuscript received:
3 April 2006
- Manuscript approved:
9 May 2006
- Version of record:
20 February 2007