Prediction of wettability variation and its impact on flow using pore- to reservoir- scale simulations Matthew Jackson, Per Valvatne and Martin Blunt Centre for Petroleum Studies Department of Earth Science and Engineering Imperial College of Science, Technology and Medicine
20
Embed
Prediction of wettability variation and its impact on flow using pore- to reservoir-scale simulations Matthew Jackson, Per Valvatne and Martin Blunt Centre.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Prediction of wettability variationand its impact on flow using pore- to
reservoir-scale simulations
Matthew Jackson, Per Valvatne and Martin Blunt
Centre for Petroleum StudiesDepartment of Earth Science and Engineering
Imperial College of Science, Technology and MedicineLondon U.K.
Impact of wettability variations
• Aim of this study is to investigate and predict the effect of wettability variations on flow at the pore- and reservoir-scales
• Use a pore-scale network model in conjunction with conventional reservoir-scale simulations
• Predict experimental relative permeability and waterflood recoveries for water-wet and mixed-wet Berea sandstone assuming wettability variations result from variations in Swi
• Predict the impact on recovery of wettability variations associated with a transition zone above the oil-water contact (variations in Swi)
The network model: detailed geometry
• 9mm3 cube containing 12349 pores and 26146 throats• Reconstructed directly from a sample of Berea sandstone
—more likely to be truly predictive
The network model: detailed physics
• Two-phase flow in layers and corners• Snap-off, piston-type displacement and co-operative pore body
filling• Allow wettability alteration after drainage by changing advancing
contact angle allocated to each oil-filled pore and throat
Drainage0r
Waterflooding 90a 90a
0.0
0.0
0.0
0.0
0.0
0.0
0.1
1.0
0.2 0.4 0.6 0.8 1.0Water Saturation
Rel
ativ
e pe
rmea
bilit
y
Prediction of relative permeability:Water-wet Berea data (Oak, 1990)
Drainage r = 0°)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.2 0.4 0.6 0.8 1.0Water Saturation
Rel
ativ
e pe
rmea
bilit
y
Imbibition a = 50-80°)Uniform distribution
Prediction of relative permeability:Water-wet Berea data (Oak, 1990)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.2 0.4 0.6 0.8
Water saturation
Rel
ativ
e pe
rmea
bilit
y
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
1.0
0.2 0.4 0.6 0.8
Water saturation
Rel
ativ
e pe
rmea
bilit
y
Prediction of waterflood recovery:Mixed-wet Berea data (Jadhunandan and Morrow, 1990)
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5Pore volumes injected
Fra
ctio
n o
f ori
gin
al o
il re
cove
red
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5Pore volumes injected
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5Pore volumes injected
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5Pore volumes injected
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°
Prediction of waterflood recovery:Mixed-wet Berea data (Jadhunandan and Morrow, 1990)
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5Pore volumes injected
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5Pore volumes injected
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5Pore volumes injected
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5Pore volumes injected
Fra
ctio
n o
f ori
gin
al o
il re
cove
red
Water-Wet Swi = 0.31 Swi = 0.24
Swi = 0.18 Swi = 0.08
a=110-180°, a=130-180°,
a=85-180°, a=110-180°
Wettability variations above OWC
z
Sw
More oil wet
Water wet
Differentinitial watersaturations
Hysteresis: Killough modelR
ela
tiv
e p
erm
ea
bil
ity
W e tt in g p h a s e s a t u r a ti o n
D ra in a g e
Im b ib i ti o n
N o n -w e tti n g
p h a s e h y s te r e s is
Re
lati
ve
pe
rme
ab
ilit
y
W e tt in g p h a s e s a tu r a ti o n
D ra in a g eIm b ib i ti o n
W e tt in g
p h a se h y s ter es is
Network model: Water-wet
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1Water saturation
Rel
ativ
e pe
rmea
bilit
y___
a = 50-80°
Network model: Oil-wet
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1Water saturation
Rel
ativ
e pe
rmea
bilit
y
a = 110-180°
Oil-wet: Killough vs. network
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1Water saturation
Rel
ativ
e pe
rmea
bilit
y
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1Water saturation
Rel
ativ
e pe
rmea
bilit
y
Killough modelNetwork model
Effect of varying initial water saturation
Swi = 0.00Sw = 0.40
Swi = 0.05Sw = 0.40
Pores contacted by oil remain water-wet
Effect of varying initial water saturation
Pores contacted by oil become oil-wet
Swi = 0.00Sw = 0.40
Swi = 0.05Sw = 0.40
Hysteresis: Effect at reservoir-scale
• Use conventional simulation to investigate effect of wettability variations on reservoir-scale flow within transition zone
• Simulate four cases:— assume reservoir is uniformly water-wet— assume reservoir is uniformly oil-wet— recognise wettability variation
— use Killough hysteresis model with oil-wet bounding curve (measured at top of reservoir)
— use relative permeability curves derived from network model
Maureen Field Simulation Model
Simulation results
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.2 0.4 0.6 0.8 1PV water injected
PV
of o
il pr
oduc
ed
Network modelWater-wet; no hysteresis
Killough modelOil-wet; no hysteresis
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1PV water injected
Wat
ercu
t
Conclusions
• Predicted experimental relative permeability and waterflood data for water-wet and mixed-wet Berea sandstone
• Emperical hysteresis models do not capture variations in relative permeability if wettability varies with height due to variations in Swi associated with capillary rise
• Relative permeabilities predicted by network model reflect pore-scale displacement mechanisms which yield low water relative permeabilities for moderate Swi
• Wettability variation has a significant effect on predicted recovery at the reservoir-scale
• Demonstrate that network models of real rocks may be used as a tool to predict wettability variations and their impact on flow at the reservoir-scale
Acknowledgements
• BHP• Enterprise Oil• Department of Trade and Industry• Gaz de France• Japan National Oil Corporation• PDVSA-Intevep• Schlumberger• Shell• Statoil