Analysis of Alternative Well-Control Methods For Dual-Density Deepwater Drilling

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

WELL INTERVENTION AND CONTROL

Analysis of Alternative Well-Control Methods


for Dual-Density Deepwater Drilling
Dual-gradient drilling methods have
been proposed to provide simpler,
safer, and more economic well designs
for deepwater gas resources. Riser gas
lift was investigated as a means to
implement a dual-gradient system. A
primary concern was whether effective
well control was possible in a system
containing so many flow paths and
fluids with different densities. A study
with a transient, multiphase simulator
found that well control was feasible by
use of methods analogous to those for
conventional operations.

Introduction
Deepwater Gulf of Mexico (GOM)
resources are especially important for
reversing the decline in U.S. production.
However, a narrow margin between
formation pore and fracture pressure
exists in many overpressured basins
including the GOM. This limited margin between pore and fracture pressure
often becomes narrower with increasing
water depth as a result of the reduced
overburden pressure and shallow onset
of abnormal pressure. As a result, reaching the target depth while retaining a
useable borehole size often is difficult,
which can limit deepwater resource
development. Dual-gradient drilling
methods have been proposed as a means
to overcome this. A dual-density drilling
concept using riser gas lift was studied
as a way to implement a dual-gradient
system. The new system would proThis article, written by Assistant
Technology Editor Karen Bybee, contains
highlights of paper SPE 98957, Analysis
of Alternative Well-Control Methods for
Dual-Density Deepwater Drilling, by M.
Stanislawek, SPE, Ensco Offshore Co.,
and J.R. Smith, SPE, Louisiana State
U., prepared for the 2006 IADC/SPE
Drilling Conference, Miami, Florida, 21
23 February.

Fig. 1Riser bottom pressure with various mud and nitrogen rates.

vide a simpler, more economic design


consisting of nitrogen and mud with an
average density equivalent to seawater
in the riser annulus and mud in the
wellbore. In concept, more standard
equipment would be used than in the
industry projects focused on use of seafloor pumps. The expected advantages
of such a system would be fewer casing
strings, larger mud-weight margins, and
a larger production-casing size.
Problem Description
The major question addressed in this
study is if effective well control is possible for a system with so many different
fluid densities and relatively complex
flow paths. Three critical phases of a
well-control operation were addressed:
kick detection, stoppage of formation
inflow, and circulation to remove kick
fluids. Each of these phases was simulated in sequence by use of a transient,
multiphase numerical simulator to determine the best well-control method.

Simulation Cases. The cases simulated in this study represent a deepwater


well in 6,000 ft of water. A relatively
high well productivity was assumed
because the reservoirs being drilled
must have a high productivity to be
economic and a high-productivity formation is more difficult to control. The
simulation assumed use of a drillstring
valve (DSV). The DSV is placed in the
drillstring near the bit to support the
excess hydrostatic pressure of the full
mud column in the drillstring when
the rig pumps are shut off. The DSV
allows mud to flow through it only
when the surface mud pumps are operating at a predetermined-setpoint pressure required to force the valve open.
Simulation was chosen to investigate
well control for a riser gas lift system
because different equipment arrangements and operating strategies can be
compared to conventional operations
without the expense of conducting fullscale experiments.

For a limited time, the full-length paper is available free to SPE members at www.spe.org/jpt. The paper has not been peer reviewed.
JPT JANUARY 2007

61

Gas Lift Feasibility


A preliminary consideration in this
study was whether dual-density drilling conditions could be achieved with
riser gas lift. For the dual-density gas
lift system to be effective, riser bottom pressure must equal the seawater
hydrostatic pressure at the mudline.
In addition, the same pressure must be
achieved at the base of the chokeline
during well-control operations with
riser gas lift.
The feasibility of achieving dualdensity conditions was confirmed by
simulating a 5,000-ft-long riser with
a 191/4-in. inside diameter (ID) and
5-in.-outside-diameter drillpipe inside
it. Various rates of 16-lbm/gal mud and
nitrogen were used. Fig. 1 shows the
results of these simulations. The bottom pressure in the riser annulus at the
mudline can be lowered successfully to
the desired seawater hydrostatic pressure. The riser circulation system with
gas injection operates in a hydrostaticdominated mode. Specifically, the large
ID of the riser limits the friction effects,
and the hydrostatic effects tend to dominate. This makes controlling the wellhead pressure straightforward because
pressure decreases for increasing gas
rate, and dual-density conditions can
be achieved at mud-circulation rates up
to 1,500 gal/min.
Kick Detection
Early kick detection minimizes kick
size and therefore decreases the difficulty of controlling the kick safely.
The kick indicators monitored during
the simulation were liquid flow rate
out, pit level, standpipe pressure, wellhead pressure, and bottomhole pressure (BHP). Dual-density drilling is
in progress in this simulation, and
after 774 minutes of drilling, a gas
kick enters the well from a formation with a 17,320-psi pressure and a
25-bbl/(psi-D) productivity index. The
first noticeable indication of a gas kick
entering the wellbore is an increase
in the return flow rate that should
be readily noticeable at the surface
under normal field conditions. Another
potential kick indicator, pit gain, may
be seen increasing with time. As the
gain increases, the indication becomes
more conclusive. Another indication
is a standpipe-pressure increase of
approximately 50 psi over a 1-minute
period and then eventually a readily
noticeable standpipe-pressure decrease

62

caused by the loss of hydrostatic pressure in the annulus as the gas volume
increases. The initial pressure peak is
the result of gas entering the well annulus and pushing the mud ahead of
it, causing additional annular friction.
Over time, hydrostatic effects tend to
dominate the whole system, and BHP
decreases significantly. These pressure
changes become conclusive only when
the pressure decrease is large and therefore are likely to be a slower indicator
than flow rate out or pit gain. For the
specific conditions examined, the kick
should be detected after no more than
3 to 6 minutes of drilling.
Stopping Formation Inflow
Two alternatives were considered for
stopping formation inflow in dual-density gas lift drilling. The first is reducing
the nitrogen rate used for riser gas lift
to increase the hydrostatic pressure
in the annulus. The second is closing
a subsea blowout preventer (BOP) to
stop flow from the well.
In a simulation where the kick was
detected and nitrogen injection shut
down 4 minutes after the gas kick
entered the well, the BHP continues to
decrease and kick volume continues
to increase. The gas kick is never controlled in this case. Given the failure to
control this kick with a reaction time
of approximately 4 minutes, shutting
down nitrogen injection is not likely to
be an effective means to control even a
moderately severe gas kick.
The second alternative for stopping
formation inflow was to close the subsea BOP, as in conventional well-control operations. For this simulation,
the rig pumps and nitrogen injection
at the seafloor were stopped at 4 minutes (778 minutes), and the BOP was
closed at 5 minutes (779 minutes),
after the gas kick began. When the
BOP is closed, BHP starts to increase
and the formation flow decreases. Flow
was essentially stopped at 790 minutes,
after 16 minutes of gas influx. The total
kick volume taken was 18.5 bbl.
A relevant issue in any well-control
situation, especially important for the
narrow margins between pore and fracture pressure, is the risk of exceeding
the fracture pressure at the casing shoe.
A specific advantage of dual-density
operations is the potential to maintain
a larger kick margin. The example presented as representative of a realistic
deepwater well in this study provides

an 800-psi kick margin at the casing


shoe for dual-density operations compared to only 200 psi for conventional
drilling. The casing-shoe pressure after
taking a gas kick and closing the subsea
BOP in dual-density drilling increases
by 335 psi, which is less than the
800-psi kick margin.
The same kick was simulated for
conventional drilling operations with
a single fluid density. The pressure at
the casing shoe increased by 259 psi,
which is above the allowed kick margin
of 200 psi. The dual-density systems
disadvantage of slightly more difficult
kick detection is offset by the advantage
of larger kick margins.
Kick Circulation
Three alternatives for circulating out a
gas kick after successfully shutting in
the well were the following.
Circulation up a gas lifted chokeline to a surface choke.
Circulation through a seafloor
choke to a gas lifted chokeline.
Circulation through a seafloor
choke into the gas lifted riser.
When formation flow stops and
BHP equals formation pressure, circulation to begin removing the kick
should begin. A special complication
is that interpreting the drillpipe-pressure buildup to determine the shut-in
drillpipe pressure (SIDPP) is not possible because of the DSV. Therefore,
a different method was adopted to
obtain an equivalent SIDPP following
pump startup.
Kick circulation in these simulations
was begun at 850 minutes. This pump
startup follows the same logic as, but
differs somewhat from, the procedure
used in conventional drilling. The
annulus pressure in the well at the seafloor, herein termed wellhead pressure,
is monitored by use of the kill line or a
pressure sensor inside the BOP stack.
The choke at the surface is opened,
pumping begins, and nitrogen injection begins. As the pump is brought to
a slow circulating rate (SCR), wellhead
pressure is kept constant by use of
choke adjustments.
Another special requirement for a
gas lift system is that the SCR must
be sufficiently high that the pump
pressure will be higher than the pressure difference between the seawater
pressure at the mudline and the mud
pressure in the drillstring at the same
depth so that there will be positive

JPT JANUARY 2007

pump pressure at the SCR. In these


cases, the SCR was 300 gal/min, and
the mud weight was kept constant at
16.0 lbm/gal. Once the SCR is reached
and pump pressure stabilizes, pump
pressure is recorded and the choke
operator switches from keeping the
wellhead pressure constant to keeping
the pump pressure constant until the
kick is removed from the well.
The first alternative considered for
circulation to remove a gas kick is
similar to the procedure routinely used
on floating rigs. The only difference is
that nitrogen is injected into the bottom of the chokeline at 7.76 MMscf/D
to reduce the hydrostatic pressure in
the chokeline and avoid lost returns.
Pump speed was kept constant, and
only choke adjustments were used to
keep drillpipe pressure and BHP constant. The nitrogen-injection rate to
the chokeline also was kept constant
to maximize the simplicity of this procedure. BHP was maintained within a
range of approximately 115 psi above
the 17,320-psi formation pressure.
Throughout the circulation period, the
wellbore pressure at the casing shoe
remained significantly less than the
9,955-psi fracture pressure. The relatively high 800-psi kick margin allowed
the gas kick to be circulated out without any risk of formation fracturing. As
a result of keeping the BHP above the
formation pressure during the gas circulation, there was no additional kick
influx, and the gas kick was successfully circulated out of the well.
The second alternative considered
would use a surface-controlled subsea
choke system to minimize the complications caused by multiphase flow in
the subsea chokeline. The advantage of
using the subsea choke is to place the
choke ahead of the multiphase flow in
the chokeline so choke pressure adjustments act more directly to affect BHP.
Kick-circulation simulation was initiated in the same manner as the first alternative. The BHP was kept within a very
narrow 45-psi range with adjustment
of the subsea choke. The advantage of
using the subsea choke instead of the
surface choke is the faster and more
accurate pressure response to choke
adjustments. The kick circulation was
successful and the risk of lost returns
even smaller than for the previous case.
The third alternative considered is to
take returns through the subsea choke
and the gas lifted riser. This would com-

JPT JANUARY 2007

bine the advantages of a subsea choke


with a means to avoid concern that friction in the chokeline could cause lost
returns. However, it would also incur
the risks of riser collapse or overpressuring the pressure-control equipment
at the top of the riser if a large gas kick
were circulated through the gas lifted
riser. Kick-circulation simulation was
initiated in a manner similar to that for
the previous alternatives. At 850minutes, the choke was opened slightly,

and mud circulation and nitrogen


injection into the riser resumed with
returns through the subsea choke and
into the gas lifted riser. Circulation of
16-lbm/gal mud down the drillstring
was increased to 500 gal/min, with pressure upstream of the choke constant
and the nitrogen-injection rate constant
at 15.52 MMscf/D. The kick was circulated out successfully, holding the BHP
within an 80-psi range with the subseaJPT
choke adjustments.

63

You might also like