ANALYSIS OF BRIGHT WATER RESERVOIR SWEEP IMPROVEMENT AND COMPARISON WITH POLYMER FLOODING FOR IMPROVED OIL RECOVERY By: AKANNI Olatokunbo Olabode THESIS Submitted in partial fulfillment of the requirements for the Degree of Master of Science in Petroleum Engineering Department of Petroleum and Natural Gas Engineering New Mexico Institute of Mining and Technology Socorro New Mexico December 2010
89
Embed
ANALYSIS OF BRIGHT WATER RESERVOIR SWEEP IMPROVEMENT …baervan.nmt.edu/publications/media/pdf/thesis/THESIS Olatokunbo... · ANALYSIS OF BRIGHT WATER RESERVOIR SWEEP IMPROVEMENT
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
ANALYSIS OF BRIGHT WATER RESERVOIR SWEEP IMPROVEMENT AND COMPARISON WITH POLYMER
FLOODING FOR IMPROVED OIL RECOVERY
By:
AKANNI Olatokunbo Olabode
THESIS
Submitted in partial fulfillment of the requirements for the
Degree of Master of Science in Petroleum Engineering
Department of Petroleum and Natural Gas Engineering
New Mexico Institute of Mining and Technology
Socorro New Mexico
December 2010
ABSTRACT
Oil recovery can be improved by injecting fluids into the reservoir via a network
of injection wells to flush oil towards the petroleum production wells. Waterflood is the
most common of this method but associated with it is the problem of early breakthrough
at the production wells and excess water production due to thief zones in the reservoir.
This study examines Bright Water reservoir sweep improvement for waterflooding and
compares with polymer flooding. The slug size of the Bright Water in the higher
permeability reservoir, the position of Bright Water slug in the reservoir and the
permeability contrast of adjacent layers are factors that affect the efficiency of this
reservoir sweep improvement method. Permeability contrast also affects oil recovery for
polymer flooding but not as much as Bright Water. The Bright Water method gives lower
recovery for highly viscous oils but maximum recovery can be obtained with polymer
flood for highly viscous oils by increasing the viscosity of the polymer to obtain
favorable mobility ratio for the displacement process. The cost relation between Bright
Water polymer and normal (HPAM) polymer also plays a role in determining the
profitability of one over the other. Early injection (before 0.5 PV) favors the Bright Water
over polymer flood, but after this the percentage of mobile oil recovered by polymer
flood passes that of Bright Water. The profitability of polymer flood is greater at early
pore volumes of injection (0.5PV – 2.0PV), and vice versa for the Bright Water treatment
method.
ii
ACKNOWLEDGMENT
First and foremost, I would like to use this opportunity to thank my research advisor, Dr.
Randy Seright. Words cannot fully capture how grateful I am to him for giving me the
opportunity to work on this project, his insightful technical knowledge and directions,
and also for his support and encouragement during the course of the research. I would
also like to recognize Dr. Her Yuan Chen and Dr. Thomas Engler for their support and
knowledgeable contributions for this project, special thanks to Dr. Thomas Engler and
Karen Balch for assisting me with unreserved access to the simulation laboratory for the
timely completion of the project. I am grateful to Dr. Robert Lee for support and the
entire staff of the New Mexico Petroleum Recovery Research Center.
I would also like to acknowledge friends and classmates who I have had the privilege of
interacting with during the course of my study, thanks to those who have contributed to
the completion of my study - directly or indirectly. Special thanks to Ronald Adegoke for
providing temporary abode for me in Socorro during my last semester. Most above all, I
will like to express gratitude to my spouse, Olufolake Odufuwa for support and
encouragement during challenging times in the course of this study.
Lastly, I dedicate this work to my parents, Dr. M.S. Akanni and Mrs. B.O. Akanni for
their continued belief in my success and labor of love to offer me the best in life. God
bless you.
iii
TABLE OF CONTENTS
ACKNOWLEDGMENT ........................................................................................................................ ii
TABLE OF CONTENTS ...................................................................................................................... iii
LIST OF TABLES ................................................................................................................................ vi
LIST OF FIGURES ............................................................................................................................. viii
APPENDIX A: Description of the Basic Reservoir Simulation Model. .............................................. 75
APPENDIX B: Description of the Polymer Model Keywords. ........................................................... 78
vi
LIST OF TABLES
TABLE 3.1: PERTINENT PROPERTIES OF THE RESERVOIR MODELS ............................................ 31
TABLE 4.1: RESULTS OF DIFFERENT SLUG POSITIONS WITH CORRESPONDING RECOVERIES
(IN %) .................................................................................................................................................. 41
TABLE 4.2: RESULTS OF DIFFERENT SLUG SIZES WITH CORRESPONDING RECOVERIES (IN
• Water viscosity: 1 cp. Other fluids (oil and polymer) viscosities are stated
when varied.
3.4 Description of Simulation Models
As stated under research objectives, the analytical results of water and polymer
flood will be compared with the results from the simulator. After this, a model for the
Bright Water treatment will be developed on this premise for the main objective of the
work; which is to compare Bright Water with polymer flood.
The software used for the reservoir simulation is Schlumberger Eclipse 300 and
part of the reservoir characteristics used in this is based on results from Kwame’s13 work
on polymer flood simulation in a heterogeneous idealized reservoir with or without
crossflow.
30
3.4.1 The Reservoir Base Case
The black-oil option is selected for the base model; a general-purpose reservoir
simulator was employed to model the performance predictions. The fully implicit
solution method was used to solve the governing equations for the simulation results
presented in this report. It includes options, which models secondary displacement and
polymer flooding for a variety of reservoir geometries. A set of grid blocks totally 50 x 1
x 2 for the xyz directions were used with each grid block sizes of 2 x 5 x 5 in meters,
respectively. Vertical communication between layers was enabled.
The injector well was located in the center of cells (1, 1, 1) and the producer was
placed in the cell (50, 1, 1) of the grid. The wells were set to perforate through both
layers in the vertical (z) direction with direct contact with the entire thickness of the
formation.
The injector wells were constrained to operate at maximum injection pressure of
78.6 atm and injection rate of 100 cubic meters per day. At the same time, the production
well was set to be constrained at bottomhole pressure of 12 atm and 100 cubic meters per
day. This calculation was made as a result of the block sizes sensitivity analysis
conducted13. Other variables including the initial reservoir conditions and PVT properties
are presented in Table 3.1.
The relative permeabilities were computed using a power law model with an
index of 2 for oil and water relative permeability curves. Water relative permeability
endpoint values of 0.1 and oil relative permeability endpoint of value of 1.0 were used.
31
Table 3.1: Pertinent properties of the reservoir models13
Reservoir thickness, m 10
Reservoir length, m 100
Permeability (k1& k2), D 0.1 & 1.0
Reservoir pressure, atm 78.6
Oil density, kg/m3 0.808264
Oil formation volume factor, rm3/sm3 1
Oil viscosities, cp 1, 10, 102, 103 and104
Oil compressibility, atm-1 0
Oil saturation, fraction 0.7
Oil production rate, m3/day 100
Water density, kg/m3 0.999125
Water compressibility, atm-1 0
Water formation volume factor, rm3/sm3 1
Water viscosity, cp 1
Initial connate water saturation, fraction 0.3
Water injection rate, m3/day 100
Number of grid blocks 50 x 1 x 2
Grid block size, m 2 x 5 x 5
Porosity, % 30
Rock compressibility, atm-1 2.0 E-8
32
The above model characteristics describe the base reservoir model used in this
work, Appendix A gives further details on the model. This base model is modified
appropriately for both the polymer flood and Bright Water cases for comparison. This
model was designed with the higher permeability layer on top of the one with lower
permeability. Tests run show a slight but negligible effect of the relative position of the
layers on oil recovery per pore volume of injected fluid. The recovery when the lower
permeability layer is on top is slightly lower than when it is below the higher
permeability layer.
3.4.2 Polymer Flood Simulation Model
The polymer option is enabled in the Eclipse simulator for polymer flood
simulation. The polymer viscosities of 10, 100, and 1000 cp were used in displacing oil
with viscosities 10, 102, 103 and 104 cp. The polymer was assigned non-Newtonian
properties to simulate and ideal solution closest to the analytical result. The injection and
the production wells were constrained at the same pressures same as that of the
waterflooding cases and also controlled by the injection and production rates as given.
Further details on the specifics of the keywords used in the simulator’s polymer option
are provided in Appendix B.
For comparison of polymer flood performance prediction plot with the analytical
results, the reservoir conditions listed above are used, but for the main comparison with
Bright Water, the fluids saturation in the layers was changed. For this second case of
comparison with Bright Water treatment; the high permeability layer was designed to be
watered out with the lower permeability layer retaining initial fluid saturation, i.e. for the
33
high permeability layer; Sw = 0.7 and So = 0.3; the low permeability layer retains initial
saturation values of Sw = 0.3, So = 0.7.
3.4.3 The Bright Water Simulation Model
The Bright Water was designed with the basic reservoir characteristics described
in the base model, but with some changes as described below:
• The higher permeability layer is watered-out with the lower permeability layer
retaining initial fluid saturations. For the high permeability layer; Sw = 0.7 and So
= 0.3. For the lower permeability layer; Sw = 0.3, So = 0.7.
• The Bright Water slug was set into position (varied for different simulation runs)
by totally blocking the pore holes (0% porosity) of the area predetermined to be
occupied by the bright Water slug. The permeability of the area to be covered is
also set to zero.
• The Bright Water slug was set in the watered out high permeability zone, with no
spillage into the lower permeability zone.
• The above assumes (optimistically) ideal behavior during the injection of the BW
into the reservoir, it assumes all the Bright Water fluid flows into the high
permeability zone blocking the only the desired (thief) zone.
The conditions described for the polymer flood and Bright Water are the base
conditions employed in the simulation for the comparison of both oil recovery methods
explained in the next chapter. Any change made would be stated before the presentation
of results. Oil viscosity of 1,000 cp is predominantly used in the comparison runs with
reason to be explained also in the next chapter.
34
CHAPTER 4
RESULTS AND DISCUSSION
This chapter presents and analyzes the simulation results of the Bright Water
profile modification process and the polymer flood recovery method. Different conditions
of both enhanced oil recovery methods are examined and comparisons are made between
the methods, also under varied conditions.
Before the presentation and analysis of the simulation results, a look is taken at
the no crossflow reservoir condition to establish the reason why simulation analysis is not
needed for the Bright Water treatment of a reservoir with no fluid flow between layers.
Then we also examine the degree of agreeability between the analytical and simulation
results for water and polymer flooding which would serve as the basis of the simulation
results comparison.
4.1 No Crossflow Reservoir Condition.
For a layered reservoir in which there is no crossflow of fluid between the layers,
there is no need to employ the sophisticated method of Bright Water treatment to improve
reservoir sweep efficiency. Since there is no crossflow between layers, once the thief
zone is blocked – at any position in the layer – there is no flow of water injected into this
high permeability layer deeper in the reservoir, implying that cheaper methods of near
wellbore treatment can be employed successfully.
Figure 4.1 shows Bright Water treatment for a reservoir with no crossflow
between layers. Applying Darcy’s law to the conditions shown in the diagram:
35
QA = QB = QC
Since Zone B is totally blocked, kB = 0, therefore QB = 0, which means QA and QC
equals zero and there is no flow in the high permeability layer.
The explanation can be applied to a near wellbore treatment for a no crossflow
condition as shown in Figure 4.2. No matter where the high permeability zone is blocked,
there is no flow in the zone, thus allowing further waterflood to properly sweep the low
permeability layer.
Figure 4.1: Bright Water treatment for a no crossflow case.
Figure 4. 2: Gel placement method for a no crossflow case.
CC
CCC
BB
BBB
AA
AAA
LPkA
LPkA
LPkA
Δ
Δ=
Δ
Δ=
Δ
Δ
***
***
***
µµµ
36
Now that it has been shown there is no need to apply a Bright Water treatment to
a no crossflow layered reservoir, the results and discussion chapter focuses on reservoir
conditions with free crossflow for the analysis and comparison of the Bright Water and
polymer flood improved recovery methods.
4.2 Validation of Simulation Results
The accuracy of the simulation results are examined before proceeding with the
presentation and analysis of results. The recovery plots obtained from the simulation is
compared with the analytical results provided by the mathematical work of Dr. Randy
Seright using fractional flow calculations as explained in the previous chapter.
The base reservoir properties and conditions explained in the previous chapter are
used in the simulation for the Bright Water and polymer flood cases; any change to the
original case is specified when made.
Figures 4.3a and 4.3b gives the waterflood recovery plots showing the
comparison of the analytical and simulation results. It is observed that the results from
the simulator generally gave higher recovery than that of the mathematical work. There is
a close match between oil with viscosities of 1 cp, 1,000 cp and 10,000 cp, but not so
with that of 10 cp and 100 cp; in which the difference considerably widens after injection
of 2 pore volumes (PV) of water. It should be noted volumetric material balance is
maintained; as the results for both the simulator and analytical method converges after
prolonged injection, validating the eventual results of the simulator.
37
Figure 4.4 shows the comparison of the analytical and simulation polymer flood
recovery plot for oil of 1,000 cp and polymer of 10 cp viscosity. This particular oil
viscosity plot is given to show the degree of agreeability between results because in the
course of the study - unless otherwise specified for further analysis – 1000 cp oil
viscosity will be used in the comparison of Bright Water and polymer flood.
Figure 4.3a: Comparison of the recovery plots from simulator and analytical method.
Figure 4.3b: Comparison of simulator/analytical method recovery plots (up to 2PV of injection)
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0 5 10 15 20
Mob
ile Oil Re
covered (frac%o
n)
Injected Pore Volume (dimensionless)
1CP Simula8on
1CP Analy8cal
10CP Simula8on
10CP Analyi8cal
100CP Simula8on
100CP Analy8cal
1,000CP Simula8on
1,000CP Analy8cal
10,000CP Simula8on
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0 0.5 1 1.5 2
Mob
ile Oil Re
covered (frac%o
n)
Injected Pore Volume (dimensionless)
1CP Simula8on
1CP Analy8cal
10CP Simula8on
10CP Analyi8cal
100CP Simula8on
100CP Analy8cal
1,000CP Simula8on
1,000CP Analy8cal
10,000CP Simula8on
38
Figure 4.4: Comparison of analytical/simulator results for 1000cp oil and 10cp polymer injected.
The discrepancy between some of the analytical and simulator results as shown by
the plots above can be attributed in part to gravity effects. As previously stated, there was
a slight change in recovery when the permeability of the layers in the model was
switched and the slight change is as a result of gravity effect. Post simulation runs also
showed that the recovery per pore volume of injected fluid decreased when the oil density
was increased to be the same as water.
The combination of the two factors above could give a better match with the
analytical and simulation results. Gravity effect is absent in fractional flow which the
analytical results are based on but this is not so for the simulator. The slight mismatch
does not affect the conclusion of this study and further work could be done to investigate
the effect of gravity on the analytical and simulation recovery plots match.
4.3 Bright Water Simulation Results
Under this section, the simulation recovery plots for different conditions of the
Bright Water treatment are examined to see which gives optimum recovery; factors such
0
0.2
0.4
0.6
0.8
1
0 2 4 6 8 10
Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volume (Dimensionless)
Simula8on
Analy8cal
39
as the position and length of the pop polymer slug in the block thief zone, change in
permeability ratio between layers is later made to see how much this affects the recovery
per pore volume of fluid injected.
4.3.1 Position of Bright Water Slug
The simulation for this case was designed for a dual layered reservoir with
conditions as specified in the previous chapter for a Bright Water case, i.e. with the
higher permeability layer is already watered out; no more mobile oil available in the
layer. The lower permeability layer retains initial fluid saturations.
The Bright Water slug was simulated for different positions in the watered out
zone to investigate the positional change effect on mobile oil recovered. The slug size
occupies 40% of this higher permeability layer; the slug was placed by the injector, in
between the injector and midpoint of the reservoir, in between the producer and midpoint,
and then by the producer with resultant plots shown in Figures 4.5 and 4.6. Figure 4.5
presents the recovery plot versus pore volumes of water injected for different positions of
the Bright Water slug. Figure 4.6 shows the same plot up to 3 PV of injection to
accentuate early recovery trend.
The plots show the early recovery - at injection below 3 PV - to be highest when
the slug is closest to the production well and this reduces as the slug’s position is moved
from the production well towards the injection well. At 3 PV of injection, the recovery
when the slug is at different positions simulated converges and after that, the recovery
when the slug is closer to the injector is higher than positions closer to the producer.
40
Table 4.1 gives the recovery values at different injection PV, for varying positions
of the Bright Water slug in the high permeability zone.
The table further shows the trend of the oil recovery with different positions of the
Bright Water slug. As stated before, the general trend shows that for injection below 3
PV there is higher recovery when the position of the slug is closer to the production well,
after that; at increased injection the recovery is higher when the position of the slug is
closer to the injection well. The 3 PV crossover point is concluded to be coincidental
since there was no reason determined for that specific pore volumes of injection.
Figure 4.5: Recovery with variation in position of 40% Bright Water slug in thief zone
0
0.2
0.4
0.6
0.8
1
0 5 10 15 20
Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volumes (Dimensionless)
From_Inj
Btw_Inj_Mid
Mid_Point
Btw_Mid_Prod
From_Prod
41
Figure 4.6: Recovery with variation in position of 40% Bright Water slug in thief zone up to 3 PV injection.
Table 4.1: Results of different slug positions with corresponding recoveries (in %)
Position of Slug From
Injector
Injector -
Midpoint
Midpoint Midpoint -
Producer
From
Producer
Recovery(%) at 0.5 PV 0.5 6.1 10.5 14.7 20.4
Recovery(%) at 1 PV 3.6 15.8 19.7 21.9 25.2
Recovery(%) at 2 PV 18.1 26.7 27.6 28.9 30.3
Recovery(%) at 3 PV 24.2 31.9 33.5 33.9 33.2
Recovery(%) at 5 PV 34.5 41.0 40.4 40.1 37.1
Recovery(%) at 10 PV 48.4 53.5 51.8 49.8 43.9
Recovery(%) at 15 PV 56.7 61.6 58.9 55.5 48.0
0
0.2
0.4
0.6
0.8
1
0 0.5 1 1.5 2 2.5 3
Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volumes (Dimensionless)
From_Inj
Btw_Inj_Mid
Mid_Point
Btw_Mid_Prod
From_Prod
42
The observed difference in the recovery as related to slug positions is explained
by Figures 4.7, 4.8 and 4.9 below. When the position of the slug is close to the producer;
early injection of water forces more oil to be pushed out through the producer faster than
when the slug is farther away from the producer (as shown in Figure 4.7), but upon
further injection the recovery becomes higher when the slug is closer to the injector. The
early high recovery when the Bright Water slug is close to the producer could also be
partly caused by gravity effect.
In the course of this study, the Bright Water slug is placed in the middle of the
high permeability layer in the reservoir when recovery results is being compared with
that of polymer flooding for a balanced optimum performance.
Figure 4.7: Slug position close to producer
Figure 4.8: Slug position in the middle
43
Figure 4.9: Slug position close to injector
4.3.2 Size of Bright Water Slug
The simulation case was designed with the same properties as that of the previous
one investigated for position of the Bright Water slug. In this case, the position of the
slug is kept constant; at the middle of the high permeability layer, and the size is varied to
observe the change in oil recovered as a result of this variation.
Figure 4.10 gives the plot of mobile oil recovered versus pore volumes of water
injected for different Bright Water slug sizes varied from zero to a hundred percent (i.e.
covering 0 to 100% of the distance in the high permeability zone).
As expected the recovery is directly proportional to the Bright Water slug size;
that is the length of the high permeability layer covered by the slug. Table 4.2 presents
the recovery values at some selected injected pore volumes of water. The recovery result
for a 100% Bright Water slug corresponded with that of a single layer reservoir of same
properties (with 40 PV of water injected). This check shows the consistency in the model.
44
Figure 4.10: Recovery with variation of Bright Water slug size in thief zone
Table 4.2: Results of different slug sizes with corresponding recoveries (in %) Slug percentage in layer (%) 20 40 60 80
Recovery (%) at 1 PV 7.9 19.5 31.4 38.3
Recovery (%) at 2 PV 18.1 27.8 37.7 46.9
Recovery (%) at 3 PV 22.8 33.6 43.4 51.8
Recovery (%) at 5 PV 33.0 41.9 51.1 60.2
Recovery (%) at 10 PV 41.1 52.2 62.9 69.9
Recovery (%) at 15 PV 48.4 58.9 69.0 76.1
The effect of permeability ratio on recovery for the Bright Water treatment
method, also with that on polymer flood is examined later in this chapter.
4.4 Bright Water versus Polymer Flood
Under this section, comparison is made between the simulation results for the
Bright Water treatment and the polymer flood methods of improved oil recovery. First we
0
0.2
0.4
0.6
0.8
1
0 5 10 15 20
Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volumes (Dimensionless)
0%
10%
20%
40%
60%
80%
100%
45
examine the recovery plots for oil of 1,000 cp viscosity for both methods and then the
plots for oil with viscosities of 1 cp, 10 cp, 100 cp and 10,000 cp are analyzed to see how
the outputs vary with different oil thicknesses.
The base reservoir conditions given in the previous chapter were used in the
simulation for both recovery methods, the recovery plot for the 1,000 cp oil - the default
oil viscosity in the base case - is analyzed first in the next subsection before looking at
the plot for the other listed oil viscosities. In the Bright Water plots, the Bright Water
volume injected before the waterflood is taken note of and included in the PV of fluid
injected.
4.4.1 Recovery comparison for 1,000 cp Oil
Figure 4.11 below shows the Bright Water - polymer flood comparison plot of
mobile oil recovered versus pore volumes of water/polymer injected for oil with 1,000 cp
viscosity. Figure 4.12 gives a closer look at the recovery comparison plot up to 3 PV on
injected fluid, and Table 4.3 presents the recovery values at different pore volumes of
both methods for quick look numerical comparison.
The Figures (4.11 and 4.12) show four curves on a plot; two each for the Bright
Water and the polymer flood simulated recovery processes. The notation PF_10cp
indicates a polymer flood of 10 cp viscosity while PF_100cp indicates a polymerflood of
100 cp viscosity for the 1,000 cp oil. The notation BW_40% indicates a Bright Water
treatment with the Bright Water slug occupying 40% of the watered out higher
permeability layer and BW_80% implies that 80% of the higher permeability layer is
46
covered by the Bright Water slug. The same notation is also used in the results given in
Table 4.3.
Figure 4.11: Bright Water versus polymer flood recovery plot for 1,000 cp oil
Figure 4.12: Bright Water vs polymer flood recovery plot for 1,000cp oil (until 3 PV injection)
0
0.2
0.4
0.6
0.8
1
0 5 10 15 20
Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volumes (Dimensionless)
PF_10cp
PF_100cp
BW_80%
BW_40%
0
0.2
0.4
0.6
0.8
1
0 0.5 1 1.5 2 2.5 3
Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volumes (Dimensionless)
PF_10cp
PF_100cp
BW_80%
BW_40%
47
Table 4.3: Comparison between Bright Water and polymer flood for 1,000 cp Oil
BW_40% PF_10 BW_80% PF_100
Recovery (%) at 1 PV 15.1 2.5 29.3 31.6
Recovery (%) at 2 PV 24.5 9.6 41.5 62.5
Recovery (%) at 3 PV 31.6 16.4 46.4 76.1
Recovery (%) at 5 PV 41.5 28.9 55.9 86.8
Recovery (%) at 10 PV 51.9 49.9 67.8 94.1
Recovery (%) at 15 PV 58.8 64.2 73.9 97.4
From the plots and table above, the Bright Water banks of 40% and 80% give a
higher recovery when compared to polymer flood of 10 cp for less than 10 PV. At 10 PV
injection, the 40% BW and 10 cp polymer both give approximately 50% recovery. After
10 PV, the recovery of 10 cp polymerflood passes that of the 40% Bright Water. The
lower recovery of the polymer flood at early injection (before 0.5 PV) can be attributed to
the time required for the polymer to displace the water resident in the watered-out higher
permeability zone. After this point the recovery picks up as more polymer is injected.
The recovery is highest for the 100 cp polymer. At 1 PV of injection, the recovery
for the 100 cp polymer is 31.6% and the next highest is that of 80% BW, which is 29.3%.
After this – 1 PV injection – the recovery from the 100 cp polymer increases at a much
higher rate than the compared method, almost doubling that of the 80% BW at 3 PV of
injection.
48
4.4.2 Recovery Comparison for Other Oil Viscosities
The recovery plots and results for other oil viscosities comparing the Bright Water
output to that of the polymer flood are given in the following figures and tables. The
basic reservoir conditions given earlier were used in the simulation cases for both
recovery methods with the oil viscosities examined.
Figure 4.13 and Table 4.4 presents the recovery comparison plot and table for 1
cp and 10 cp oil, Figure 4.14 and Table 4.5 gives that for 100 cp oil, and the comparison
results for 10,000 cp oil are presented in Figure 4.15 and Table 4.6.
Figure 4.13: Bright Water versus polymer flood recovery plot for 1 cp and 10 cp oil
0
0.2
0.4
0.6
0.8
1
0 1 2 3 4
Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volumes (Dimensionless)
1cpOil_40%BW
1cpOil_1cpWater
10cpOil_40%BW
10cpOil_10cpPolymer
49
Table 4.4: Comparison between Bright Water and polymer flood for 1 cp and 10 cp oil 1cpOil
BW_40%
1cpOil
1cpWater
10cpOil
BW_40%
10cpOil
10cpWater
Recovery (%) at 0.5 PV 95.9 85.4 67.9 41.7
Recovery (%) at 1 PV 99.9 93.9 84 86.1
Recovery (%) at 2 PV 99.9 98.8 92.9 97.5
Recovery (%) at 3 PV 99.9 99.6 95.5 98.9
Recovery (%) at 3.5 PV 99.9 99.9
97.6 99.5
Figure 4.14: Bright Water versus polymer flood recovery plot for 100 cp oil
0
0.2
0.4
0.6
0.8
1
0 2 4 6 8 10
Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volumes
40%BW
80%BW
10cpPolymer
100cpPolymer
50
Table 4.5: Comparison between Bright Water and polymer flood for 100 cp oil BW_40% PF_10 BW_80% PF_100
Recovery (%) at 0.5 PV 30.4 9.5 33.3 24.8
Recovery (%) at 1 PV 48.9 39.1 63.5 75.9
Recovery (%) at 2 PV 62.3 64.5 77.1 95.6
Recovery (%) at 3 PV 69.7 75.3 83.6 99.2
Recovery (%) at 5 PV 77.0 85.7 87.8 99.9
Recovery (%) at 10 PV 87.1 93.8 93.9 99.9
Figure 4.15: Bright Water versus polymer flood recovery plot for 10,000 cp oil
0
0.2
0.4
0.6
0.8
1
0 2 4 6 8 10
Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volumes (Dimensionless)
40%BW
80%BW
100cpPolymer
1,000cpPolymer
51
Table 4.6: Comparison between Bright Water and polymer flood for 10,000 cp oil BW_40% PF_100 BW_80% PF_1,000
Recovery (%) at 1 PV 1.8 1.9 12.1 36.2
Recovery (%) at 2 PV 2.7 6.4 18.7 67.3
Recovery (%) at 3 PV 5.1 12.5 23.9 78.2
Recovery (%) at 5 PV 10.8 23.6 29.8 87.9
Recovery (%) at 10 PV 19.2 48.8 38.9 94.8
For 1 cp oil; it is observed that the 40% Bright Water treatment and the
waterflood give approximately the same recovery of mobile oil after 1 PV of water
injection, as expected the Bright Water treatment gives higher recovery prior to the 1 PV
injection point.
For 10 cp oil; at 0.5 PV of injection the recovery for 40% Bright Water was
higher than the 10 cp polymer flood but this is reversed at 1 PV of injection when the
polymer flood recovery is 2.1% higher than the Bright Water method. Again, as
expected, Bright Water gives a higher recovery than polymer flood at early injection.
For 100 cp oil; the 40% and 80% Bright Water give higher recoveries than 10 cp
and 100 cp polymer flood at 0.5 PV of injected fluids. This changed after 1 PV of
injection when the recovery from 10 cp polymer is higher than 40% Bright Water but still
lower than 80% Bright Water and the recovery from 100 cp is the highest.
For 10,000 cp oil; at 1 PV injection the 1,000 cp polymer gives a far higher
recovery than the Bright Water at 80% and the 100 cp polymer gives approximately the
52
same recovery as the 40% Bright Water. Continued injection maintained this trend, and
the recovery for 100 cp polymer passes that from 80% Bright Water at 7 PV of injection.
From the above results, the polymer flood method has the capability to give
higher recovery than Bright Water. The higher the oil viscosity, the higher the recovery
of polymer floods over Bright Water because the viscosity of the polymer can be
increased to improve recovery. Bright Water gives better recovery at early injection
stages.
4.5 Permeability Ratio
In this section, the effect of permeability ratio between the layers on mobile oil
recovered is examined for both the Bright Water treatment and the polymer flood
method.
4.5.1 Permeability Ratio Effect on Bright Water Recovery
The base reservoir conditions previously listed are used in the simulation to
examine the effect of permeability ratio on the Bright Water treatment recovery. The
permeability ratio is varied for three different cases; with oil viscosity of 1,000 cp. 40%
of the high permeability layer was covered by the Bright Water slug (40% BW).
Figure 4.16 below shows the recovery comparison plot for the different
permeability ratios of 2:1, 5:1 and 20:1 of a Bright Water treatment model, Table 4.7
provides the recovery values per pore volume of water injected. As seen from the plot
and table of result, only a slight difference was observed in the mobile oil recovered with
different permeability ratios for the Bright Water treatment method.
53
Figure 4.16: Permeability ratio comparison plot for Bright Water (Base reservoir conditions)
Table 4.7: Permeability ratio comparison for Bright Water (Base reservoir conditions) Permeability Ratio 2:1 5:1 20:1
Recovery (%) at 1 PV 20.9 18.6 17.5
Recovery (%) at 4 PV 40.0 38.9 37.1
Recovery (%) at 8 PV 52.7 50.0 48.5
In order to accentuate the effect of the permeability ratio on recovery for Bright
Water, the simulation base case was modified from 2 grid blocks in the vertical direction
to 6 grid blocks; other conditions were kept the same. Fig. 4.17 presents the result of this
grid block modification.
The plot shows the recovery of the model with 6 grid blocks in the vertical
direction and the base one with 2 grid blocks in the vertical direction to be the same and
the refining of grid blocks had no effect on the result. This confirms the permeability
ratio only has a slight effect on the recovery for the 40% Bright Water.
0
0.2
0.4
0.6
0.8
1
0 2 4 6 8 10 Mob
ile Oil Re
covered (Frac%on
)
Injected Pore Volumes (Dimensionless)
2_1
5_1
20_1
54
Figure 4.17: Permeability ratio comparison plot for Bright Water (6 grid blocks in vertical axis) The percentage of high permeability layer covered by the Bright Water slug is
reduced from 40% to 10%, with other conditions kept constant, then the simulation is run
to obtain the Bright Water recovery plot. Figure 4.18 and Table 4.8 present the result for
the modified case of 10% Bright Water.
The results from the plot and table show the effect of the permeability ratio on
mobile oil recovered is more pronounced with lower Bright Water slug size. The results
show that with reduced permeability ratio, the recovery is slightly higher.
Figure 4.18: Permeability ratio comparison plot for Bright Water (with 10%BW)
0
0.2
0.4
0.6
0.8
1
0 2 4 6 8 10 Mob
ile Oil reciovered
(Frac%on
)
Injected Pore Volumes (DImensionless)
2_1
5_1
20_1
0
0.2
0.4
0.6
0.8
1
0 2 4 6 8 10 Mob
ile Oil reciovered
(Frac%on
)
Injected Pore Volumes (DImensionless)
2_1
5_1
20_1
55
Table 4.8: Permeability ratio comparison for Bright Water (with 10% BW) Permeability Ratio 2:1 5:1 20:1
Recovery (%) at 1 PV 5.3 2.9 0.9
Recovery (%) at 4 PV 25.1 20.1 16.8
Recovery (%) at 8 PV 39.2 31.1 27.8
4.5.2 Permeability Ratio Effect on Polymer Flood Recovery
To examine the effect of permeability ratio on polymer flood recovery, the base
reservoir conditions previously used for the polymer flood base model was employed
with the permeability ratio varied for three different cases; 2:1, 5:1 and 20:1. Oil viscosity
of 1,000 cp is used. Figure 4.19 and Table 4.9 presents the result for the permeability
ratio comparison for the polymer simulation case described above with 10 cp polymer.
The result shows a considerable difference in recovery for different permeability
ratios, and the recovery is higher for reduced permeability ratio as observed in the plot in
Figure 4.19.
Figure 4.19: Permeability ratio comparison plot for polymer flood (with 10 cp polymer)
0
0.2
0.4
0.6
0.8
1
0 2 4 6 8 10 Mob
ile Oil recovered (Frac%on
)
Injected Pore Volumes (Dimensionless)
2_1
5_1
20_1
56
Table 4.9: Permeability ratio comparison for polymer flood (10 cp polymer)
Permeability Ratio 2:1 5:1 20:1
Recovery (%) at 1 PV 9.1 4.2 1.1
Recovery (%) at 4 PV 60.0 37.9 11.0
Recovery (%) at 8 PV 76.2 61.1 25.3
4.5.3 Permeability Ratio Effect Comparison of Bright Water and Polymer Flood
The plots of the simulation results obtained in the previous subsections for the
permeability effect on recovery for the Bright Water and polymer flood are combined to
view the difference of the effect on both methods.
Figure 4.20 shows the permeability ratio effect comparison plot for Bright Water
and Polymerflood. For the Bright Water, 40% of the higher permeability layer is covered
by the Bright Water slug and a 10 cp polymer is used in the polymer flood.
The figure shows how much effect the permeability ratio has on the recovery for
both methods, the recovery is higher for lower permeability ratios and the effect is greater
in polymer flood than the Bright Water.
Figure 4.20: Comparison plot for permeability ratio effect on polymer flood and Bright Water
0
0.2
0.4
0.6
0.8
1
0 2 4 6 8 10
Mob
ile Oil recovered
(Frac%on
)
Injected Pore Volumes (Dimensionless)
2_1_10cp
5_1_10cp
20_1_10cp
2_1_40%BW
5_1_40%BW
20_1_40%BW
57
CHAPTER 5
ECONOMICS CONSIDERATIONS
For the conclusion of this study, financial comparison is made between polymer
flood and the Bright Water treatment methods. A basic form of cost comparison is
employed in order to determine how the costs relate under different conditions. The price
of normal polymer; hydrolyzed polyacrylamide (HPAM) is between $0.9/lb to $2.0/lb15,
and the Bright Water polymer costs around 5 times as much as that for HPAM
(discussion with R.S. Seright, August 2010). Three price situations are considered for the
normal polymer and Bright Water polymer costs comparison:
i. Normal Case: Bright Water polymer costs five times as much as normal polymer.
ii. Optimistic Case: Bright Water polymer costs twice as much as normal polymer.
iii. Extremely Optimistic Case: Bright Water polymer costs the same as normal
polymer.
5.1 Bright Water Polymer Concentration
The concentration for the Bright Water polymer in injected water is 10,000 ppm
(i.e. 1%), and since the Bright Water injection is performed once in the operation - before
further water injection - the cost associated with the two major amounts of Bright Water
used in the course of this research is calculated once.
The two major Bright Water cases examined in this research are; when 40% of
high permeability layer is covered by Bright Water, and when 80% of high permeability
layer is covered by Bright Water.
58
5.1.1 Bright Water for 40% of High Permeability Layer
From the simulation model used in this study, the pore volume of each layer is
750 cubic meters, so 40% of the high permeability layer equals 300 cubic meters. This is
based on the optimistic assumption that all the Bright Water injected goes directly into
the high permeability layer as intended. The mass of Bright Water polymer involved (by
weight) can be calculated thus:
Bright Water polymer mass in 40% injection = kgmmkg 3000300001.010000 3
3 =××
5.1.2 Bright Water for 80% of High Permeability Layer
80% of the high permeability layer is 600 cubic meters. This is again based on the
optimistic assumption that all the Bright Water injected goes directly into the high
permeability layer as intended. The mass of Bright Water polymer involved can be
calculated thus:
Bright Water polymer mass in 80% injection kgmmkg 6000600001.010000 3
3 =××
Three price situations are to be examined regarding the relation of the cost of
Bright Water polymer and normal (HPAM) polymer.
59
5.2 Polymer (HPAM) Concentration
As given in the simulation model for the Bright Water treatment case, the pore
volume for each layer is 750 cubic meters, thus the total pore volume of the reservoir
model is 1500 cubic meters. The concentration of different polymer viscosities is
calculated based on the following HPAM polymer concentration in parts per million15:
10 cp = 900 ppm
100 cp = 3,000 ppm
1,000 cp = 10,000 ppm
10 cp Polymer
10 cp polymer mass in 1 PV kgmmkg 13501500001.0900 3
3 =××=
2 PV = 2700 kg, 3 PV = 4050 kg
100 cp Polymer
100 cp polymer mass in 1 PV kgmmkg 45001500001.03000 3
3 =××=
2 PV = 9000 kg, 3 PV =13500 kg
1000 cp Polymer
1000 cp polymer mass in 1 PV kgmmkg 150001500001.010000 3
3 =××=
2 PV = 30000 kg, 3 PV = 45000 kg
60
5.3 Cost Comparison for Bright Water and Polymer Flood
The cost of Bright Water polymer used is compared to that of normal polymer at
different pore volumes of polymer injected and for different oil and polymer viscosities
based on the recovery results from the simulator. The effect of the cost of water handling
for both processes is neglected as this is assumed approximately equal for both processes.
This comparison is made for oil of 10 cp, 100 cp, 1,000 cp and 10,000 cp
viscosities. This is done for different situations when the Bright Water polymer costs five
times as much as normal polymer, twice as much as normal polymer and the same as
normal polymer.
5.3.1 Normal Case Comparison
For this case, the Bright Water Polymer costs five times as much as the normal
polymer. The normal (HPAM) polymer costs $3.3 per kilogram and the Bright Water
polymer costs five times as much as this, which is $16.5 per kilogram.
For the comparison, a benefit ratio index is calculated by the relation below:
Benefit ratio = (percentage of oil recovered) / (cost of polymer or Bright Water injected)
Cost of normal polymer = Amount of polymer (kg) x $3.3/kg
Cost of Bright Water polymer = Amount of Bright Water polymer (kg) x $16.5/kg
Tables 5.1 through to 5.4 present the results the benefit ratio for cost comparison
of different oil viscosities. The comparison is made up to 3 PV of injection. The recovery
values are obtained from the tables of results in the previous chapter.
61
Table 5.1: Benefit ratio (i) for cost comparison of 10 cp Oil 40% BW
Recovery (%)
40% BW
Benefit Ratio
10 cp polymer
Recovery (%)
10 cp polymer
Benefit Ratio
1 PV 84 1.69E -3 86.1 1.93E -2
2 PV 92.9 1.88E -3 97.5 1.09E -2
3 PV 95.5 1.92E -3 98.9 7.39E -3
Table 5.2: Benefit ratio (i) for cost comparison of 100 cp Oil 40% BW