The Benefits and Risks of Fractures in Enhanced Oil Recovery R.S.

Transcription

The Benefits and Risks of Fractures in Enhanced Oil Recovery R.S.
The Benefits and Risks of
Fractures in Enhanced Oil
Recovery
R.S. Seright,
New Mexico Tech
RISKS ASSOCIATED WITH FRACTURES IN EOR
v Fractures may cause direct channeling
between injection wells and production wells.
v Fractures may extend “out of zone”.
Sometimes, we overlook the effects of fractures
on sweep efficiency—e.g., CO2 flooding.
To reduce channeling during CO2 floods, most
people think of WAG or foams.
However, in fractures, WAG and foams have
limited effectiveness.
Most CO2 floods occur in 1-10 md carbonates,
where many natural fractures exist.
Permeability of a 1-mm-wide fracture is over 8
million times greater than that for 10-md rock.
Fractures can also have a very positive impact
on waterfloods and EOR projects.
For water and surfactant imbibition processes, large
fracture areas are critical to making the process work.
FRACTURES IN POLYMER FLOODING
1.  With vertical wells, fractures or fracture-like features
must be open during polymer injection.
2.  Simple radial flow equation (polymer/water
injectivity):
I /Io = Ln (re/rw) / [Fr Ln (rp/rw) + Ln (re/rp)]
Assume re=1000 m, rw=0.1 m, rp = 100 m.
Fr
I /Io
3
0.40
10
0.13
20
0.07
50
0.03
3. Oil producers will not tolerate large injectivity losses
—because reduced injection rate means reduced oil
production rate.
5
Even without face plugging, the viscous nature of
polymer solutions requires that injectivity must
be less than 20% that of water if formation parting
is to be avoided.
Injectivity relative to water
1
1
Vertical well, 20-acre 5-spot, φ =0.2
3 cp Newtonian
HPAM: Fr = 3.7 + u2/1960
10 cp Newtonian
0.1
0.1
xanthan: Fr = 2.5 + 20 u-0.5
30 cp Newtonian
100 cp Newtonian
HPAM: Fr = 42 + 11 u
0.01
0.01
0
0.1
0.2
0.3
0.4
0.5
PV injected
6
Fractures in Polymer Flooding
Injection has occurred above the formation
parting pressure for the vast majority of polymer
floods.
Fractures simply extend to accommodate the rate
and viscosity of the fluid injected.
So injectivity is rarely a problem for polymer
floods unless a pressure constraint is imposed.
What is a reasonable pressure constraint? How
far is too far for fracture extension?
Increasing fracture length to 30% of the total interwell
distance reduces sweep efficiency from 0.63 to 0.53.
Increasing polymer viscosity from 10 to 100 cp
increases recovery from 0.16 to 0.54.
Mobile oil recovered at 1 PV
1
1000-cp oil. 2-layers with crossflow.
k1=10k2. h1=h2. 5-spot pattern. Fracture
points directly at producer. Assumes all
oil within 1 fracture radius from injector
is bypassed.
0.9
0.8
0.7
Polymer
viscosity
10 cp
20 cp
33.3 cp
50 cp
100 cp
0.6
0.5
0.4
0.3
0.2
0.1
0
0
0.1
0.2
0.3
0.4
Fracture length relative to injector-producer distance
0.5
Injectivity and Fracture Extension
Tambaredjo Field (Suriname), Moe Soe Let et al. (2012):
horizontal fractures extended <30 ft from the injection
well (well spacing was 300 ft).
Matzen Field (Austria), Zechner et al. (2015): vertical
fractures only extended 43 ft from the injection well (well
spacing was 650-1000 ft).
No problems were reported with injectivity, or of
fractures compromising the reservoir seals or causing
severe channeling during the Daqing project (Han 2015),
even injecting 150-300-cp polymer.
With no fracture, injectivity, productivity,
and pressure gradients are low for most of
a 5-spot pattern.
20
18
Pressure, MPa
16
14
12
10
8
6
4
2
150
0
0
75
75
150
x, m
225
y, m
0
300
10
With fractures open near injectors and
producers, injectivity, productivity, and
pressure gradients are high—even if the
fractures point directly between the wells.
20
18
Pressure, MPa
16
14
12
10
8
6
4
2
150
0
0
75
75
150
x, m
225
y, m
0
300
11
FRACTURES: BOTTOM LINE
1. Fractures can be bad or good (even essential) for EOR.
2. For most previous polymer floods, injection has
occurred above the formation parting (fracture) pressure
—even though the operators insisted that they did not.
3. This is not bad, so long as fracture extension is
controlled so that fractures don’t (a) let fluids “flow out
of zone” or (b) extend far enough to cause channeling.
4. Be realistic. If you can’t live with the injectivity
reduction associated with a viscous fluid, don’t insist
that you are going to inject below the parting pressure.
5. If you are willing to inject above the parting pressure,
spend some time to understand how the fractures will
extend and the consequences.
12
Even with the cleanest polymers, face plugging
will exceed the capacity of unfractured wells
during most chemical EOR projects.
10000
Throughput, cm 3/cm2
3000
cm3/cm2
1000
600
cm3/cm2
100
cm3/cm2
100
10
1
0.0001
20-ac spacing,
rw=0.375 ft, φ = 0.2
0.001
0.01
PV injected
0.1
1
14
Scheme to Maximize Polymer Injectivity/Productivity
Horizontal
Injector
Injector
Fractures
Minimum
stress
direction
Horizontal
Producer
Producer
Fractures
15
A DILEMMA FOR POLYMER FLOODING
1. Injecting above the parting pressure is often necessary
for adequate injectivity.
2. If polymer breaks through early, how can you tell if it is
because of a fracture or viscous fingering?
3. If breakthrough occurs from a fracture, you should
decrease the injection rate and/or polymer viscosity.
4. If breakthrough occurs from viscous fingering, you
should increase the polymer viscosity.
• Transit through fractures that cause severe channeling
should occur fast—days or less.
• Transit through viscous fingers typically takes months.
16
Simulation of Polymer Injectivity:
Assuming two “wrongs” to try to make a “right”
Several simulators assumed (1) injectors are not
fractured and (2) HPAM solutions show shearthinning behavior at near-wellbore velocities.
They claim to match injectivity behavior, but both
assumptions are wrong.
By incorrectly assuming no fractures are present,
the simulations predict a false (low) “economic”
optimum polymer viscosity.