Lee - Effect of water injection above fracturing condition

Copyright

by

Kyung Haeng Lee

2008

IMPACT OF INJECTION WELL FRACTURES ON WELL

INJECTIVITY AND RESERVOIR SWEEP IN

WATERFLOODING AND ENHANCED OIL RECOVERY

by

Kyung Haeng Lee

REPORT

Presented to the Faculty of the Graduate School

of The University of Texas at Austin

in Partial Fulfillment

of the Requirements

for the Degree of

MASTER OF SCIENCE IN ENGINEERING

The University of Texas at Austin

May 2008




APPROVED BY
SUPERVISING COMMITTEE:

__________________________

Mukul M. Sharma

__________________________

Chun Huh

ACKNOWLEDGEMENTS

I would like to appreciate my parents, Jaipoong Lee and Myungsook Kang,

and my parents in law, Samjun Bae and Youngsook Jung for the endless support.

Without them, I could not complete my Master degree. Also, I would like to express

my gratitude and love to my wife, Seunghyun Bae and daughter, Cheny Lee for being

with me, encouraging and helping me to concentrate only on the study.

I sincerely thank Dr. Sharma for the advising and encouragement that he has

offered throughout the graduate years. Working with him was truly an inspiration. I

deeply express my gratitude to Dr. Huh for the continuous support and guidance to

the right path. Also, special thanks to Mr. Choi for helping me to adapt myself to the

research during my first semester.

I would like to show my gratitude to Joanna L. Castillo who supports software

to carry out my simulation work. Also, I am grateful to Dr. Tergian for sincerely

supporting my research environment.

Finally, I would like to attribute this glory to my parents, parents-in-law in

South Korea. I always appreciate their unconditional love and continuous support.

iv

ABSTRACT




by

Kyung Haeng Lee, M.S.E.

The University of Texas at Austin, 2008

SUPERVISOR: Mukul M. Sharma

Water injection is widely used to maintain reservoir pressure and to displace

bypassed oil from unswept zones. During the water injection process, deposition of

suspended solids and oil droplets at the wellbore zone leads to a decline in well

injectivity. Hence, an increased injection pressure is required to maintain a given

injection rate. If the increase in the injection pressure is such that it exceeds the

minimum horizontal stress within the formation rock around the wellbore, fractures

are initiated in the adjacent formation. If the temperature of the injected fluid is

different from that of the formation, a thermal front propagates from the injection

v

well. This change in temperature causes the rock to contract or expand, thereby

altering the stresses both in the region of changed temperature and in the surrounding

rock. For example, injection of cold water into a high temperature reservoir can

induce thermal stresses in the near wellbore region, which facilitates fracturing. The

above two processes, pore plugging and changes in the temperature of the rock, are

the main mechanisms that drive injection well fractures.

To maximize the oil recovery the consideration of fracture growth rate and

fracture orientation is essential. The extent of fracture growth and the fracture

orientation significantly affects the sweep efficiency for given well pattern.

Therefore, in the reservoir with complicated well patterns, the optimum fracture

growth rate and fracture orientation is essential in maximizing the oil recovery. The

appropriate selection of injection rate and the knowledge of particle concentration of

the water and the temperature of the water are key factors necessary to determine the

optimum fracture growth rate.

Therefore, the accurate oil recovery simulation should include the detailed

description of the fracture growth during the water injection. However, there is no

reservoir simulator which explicitly considers fracture growth during the simulation

so far. The usual simulator considers just the fixed fracture in the reservoir. However,

because the fracture grows continuously as the injection of water progresses, a proper

consideration of the fracture growth is necessary in the process of reservoir

simulation.

vi

To add the explicit accounting of the fracture growth to the reservoir simulator

will be the initial focus of my thesis research, as I describe it in the method for

conducting research. After completing the combination of two simulators (single-well

fracture growth simulator and reservoir simulator), the optimum fracture growth rate,

optimum fracture orientation and optimum properties of injected water can be

obtained by a sensitivity study.

vii

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ................................................................................ iv

ABSTRACT........................................................................................................... v

TABLE OF CONTENTS .................................................................................. viii

LIST OF FIGURES ............................................................................................. ix

1. Research Objective ........................................................................................... 1

2. Research of Relevant Literature………………...............................................3

3. Method for Conducting the Research ............................................................. 9

4. Preliminary Results ........................................................................................ 15

5. References........................................................................................................ 20

VITA..................................................................................................................... 22

viii

LIST OF FIGURES

Figure 1: The detailed steps for combining two simulators 11

Figure 2-A: Five spot injection pattern with unfavorable direction of fracture 15

Figure 2-B: Five spot injection pattern with favorable direction of fracture 16

Figure 3: Effect of injection rate on dimensionless production 17

Figure 4: Effect of well spacing on dimensionless production 17

Figure 5: Effect of fracture length and mobility ratio on dimensionless

production 18

Figure 6: Effect of fracture length on dimensionless production 19

ix

1. Research Objective

The primary objective of this research is to determine the impact of fractures

and complex wells on oil recovery achieved by water flooding and Enhanced Oil

Recovery (EOR).

This research objective will be met by simulating the growth of fractures in

water injection wells as well as polymer injection wells. The approach that we intend

to follow is to develop analytical and numerical models for fracture growth in

injectors. We then proposed to combine single well models (UTWID) with reservoir

simulator (UTCHEM or GEM) so that the physics of fracture growth can be modeled

accurately in the single well models while the reservoir architecture can be

incorporated in the reservoir simulator. Models will also be developed for deviated

and horizontal wells. These models will incorporate the effects of complex fluid

rheology (shear thinning and viscoelastic effects) as well as thermal stresses induced

by fluid injection and particle plugging effects due to solids and oil droplets in the

injected fluids.

The results of this research will provide clear understanding of the impact of

both hydraulic fractures and dynamically growing injection well fractures in both

producers and injectors. Simulations will also help to identify conditions under which

1

complex wells and fractures may be used to accelerate oil production in water

flooding and EOR processes.

2

2. Review of Relevant Literature

The effect of hydraulic fractures of constant length on oil recovery and

injectivity has been investigated in detail in the past. Improvements in injectivity of a

factor of 2 to 5 are computed and observed in the field. Wang et al. (1970) showed

that the creation of fractures in injectors can accelerate oil production but can also

lead to early water breakthrough in water flooding applications.

In 2001, Gadde et al. combined single injection well simulator with reservoir

simulator and for the first time simulated oil displacement efficiency in systems

where the fracture length was increasing with time. The physics of fracture growth

was modeled consistently by taking into account thermal stresses as well as particle

plugging effects. Single well injection model (UTWID) was combined with a three

dimensional chemical flooding simulator (UTCHEM) to capture the effects of

fracture growth rate, fracture orientation and reservoir properties on oil recovery. This

study was designed to investigate the impact of growing injection well fractures on

water flood performance and reservoir sweep. Deviated or horizontal wells were not

considered. The Injection of reologically complex fluids such as polymers was also

not included in this investigation. In addition, the study was limited to water flooding

applications.

3

Pang et al. (1997) investigated the modeling of the injectivity decline in water

injection wells. The reason for fracture generation is the injectivity decline due to the

particle and oil droplet in the water. The fracture growth rate and its relation with

injection rate and the concentration of the injected water are shown in this paper.

Also, the concept of the transition time was proposed in this paper at first. The

transition time means the time at which no more particles invade the rock, the time at

which the initial layer of external filter cake is completely formed. The concept of

transition time makes the filtration models suggest both internal and external filter

cakes for the first time.

Barkman et al. (1972) proposed the water quality ratio which is the

concentration of suspended solids to the permeability of the filter cake formed by

those solids. It can be used to calculate the rate of formation impairment.

Perkins and Gonzalez (1985) investigated the mechanism behind how the

temperature difference between injected water and reservoir should affect the

thermoelastic properties in the reservoir and how it affects the fracture growth as

water is injected. For typical water flooding of a moderately deep reservoir,

horizontal earth stresses may be reduced by several hundred psi.

Peaceman (1983) investigated the interpretation of well-block pressures in

numerical reservoir simulation with anisotropic permeability. The interpretation of
4

well-block pressures could be progressed even though the grid blocks were not square

and the permeability distribution was not isotropic.

Stenebraten et al. (2002) investigated experimentally the growth of the

injection well fractures using large block tests. They showed clearly that the growth

of the fracture was closely related to particle plugging of the fracture face. The

injection of plugging particles significantly increases the fracture growth rate. In

periods of clean water injection, the fracture does not grow. Most of the injected

particles were found to be trapped within a very short distance from the face of the

fracture and near the tip of the fracture. Despite of plugging of the fracture face by

particles, the injectivity remains constant over long periods of time.

Wang et al. (2007) established the optimum polymer formulations, injection

rates, and individual well production allocations, and time-dependent variation of the

molecular weight of the polymer used in the injection slugs. Because of the economic

problem, the optimal time to change frac-fluid from polymer to water or low

molecular weight polymer should be determined.

In summary, particles in the water injection can plug in the fracture face

during water injection. Hence, an increased bottom hole injection pressure is required

to maintain a given injection rate. Due to the increased bottomhole pressure, a

fracture can be generated in the formation.
5

Also, the temperature of the injected water affects the hydraulic fracturing.

Usually, the temperature of the injected water is lower than the temperature of the

formation. Due to the temperature difference between the injected water and the

formation, the horizontal earth stresses may be reduced by several hundred psi. With

injecting water and particles in the water, hydraulic fracturing generates easily as a

result of reducing horizontal earth stresses. This is termed ‘thermally induced

fracturing’.

Fracture orientation and fracture growth rate play important roles in

determining the maximum production rate and ultimate oil recovery. Growing

fractures from the injection well in the direction of the producer results in premature

water breakthrough. On the other hand, the fracture toward the space between two

producers makes the sweep efficiency higher. In other words, the orientation of the

fracture growth may result in different sweep efficiency depending upon the

orientation of the fracture relative to the position of the injection wells.

To show the effect of the orientation of injection well fractures, two different

well positions can be considered. The first case is when the injection well fractures

are growing directly towards the producers. The second case is when the orientation

of the injection well fracture lies between the producers. The second case in which

the orientation of the injection well fracture lies between the producers is better

because the first case results in poor sweep and premature water breakthrough. This
6

means that the water could not sweep the oil in the reservoir and there is a substantial

amount of remaining oil in the reservoir after water flooding.

Whether fracture growth rate increases the oil recovery or not depends on the

fracture orientation. In the case of favorable fracture orientation, which means the

fracture lies between the producers, the poor sweep area by the flood front results in

lower oil recovery due to slow fracture growth. As fracture growth rate increases, the

ultimate oil recovery is increased. On the other hand, there is unfavorable fracture

orientation case, which means the fractures are growing directly towards the

producers. In the case of really slow fracture growth, the situation is the same as the

favorable fracture orientation case. However, too high fracture growth results in

premature water breakthrough, the worst case.

To date, no fracture simulator has considered dynamically growing fractures

in EOR processes. This research proposal aims to develop a simulator for fracture

growth and injection wells and combine it with the reservoir simulator to predict the

performance of EOR processes and water floods in reservoirs that contain fracture

injectors and producers. The effective important parameters such as water quality,

water temperature, reservoir properties, injection and production well patterns,

orientation of fractures, and the placement of horizontal or deviated wells will be

considered in the simulations. These simulations will help us to devise optimum

injection schemes and well placement strategies for water flooding and EOR
7

processes so that the time to first oil can be reduced and reservoir sweep can be

maximized.

8

3. Method for Conducting the Research

Task1: Impact of static fractures on reservoir sweep and injectivity

The first set of simulations that we proposed to conduct are in simple reservoir

patterns such as a 5 spot, a line drive and an inverted 5 spot with both unfractured and

fractured injectors and producers. In these simulations the fracture length will be held

constant with time. The oil recovery versus time and the injectivity will be compared

for the different cases as a function of fracture orientation, fracture length and

reservoir properties.

Task 2: Derivation of analytical and numerical models

Analytical and numerical models will be derived for fracture growth in

polymer injection wells. These models will be based on earlier models developed for

water injection wells for vertical and horizontal wells. It is expected that the polymer

rheology including shear thinning and viscoelastic behavior will play an important

role in these models and simulation results. The models will include residual

resistance factors for the polymer as well as particle plugging and thermal effects that

may be important. The model will be implemented in a numerical single well model,

UTWID that currently simulates water injection into vertical and horizontal injectors.

9

Task 3: Combining a single well model for injectors with reservoir simulator

Single well injector model that accounts for fracture growth will be combined

with a reservoir simulator (UTCHEM 9.9 and GEM). The proposed strategy for

combining these two models is shown in figure 1. Key variables from the reservoir

simulator such as simulation time, average reservoir pressure, and reservoir properties

will be transmitted to the single well injector model which will then compute the

fracture length based on the physics of fracture growth for both water injection and

polymer injection. The computed fracture length will be used to repopulate the

transmissibility matrix for the reservoir simulator so that the new fracture length can

be incorporated into the pressure profile and sweep calculations. Similar approach

was followed by Gadde (2001) for water injection into vertical wells. Incorporating

more complex well geometries, such as horizontal wells and polymer injections into

this scheme will pose some challenges.

10

START

GEM reads it input from gem.dat
GEM calls UTWID which reads its
input from utwid.dat

N=1

N=N+1

Run GEM for timestep N

GEM calls UTWID

UTWID
• Gets GEM’s n level variables
• Performs UTWID own calculations
• Write its output to utwid.out
INTERFACE
By using fracture length, calculates
Time=end time
Permeability and porosity for timestep N
Makes the new input file for GEM
STOP

Figure 1: The detailed steps for combining two simulators

Combining two simulators in this manner can be accomplished not only for

homogeneous reservoir, but also for heterogeneous reservoir. A reservoir with

heterogeneous permeability and porosity distribution could have different fracture

half lengths in different layers or on the opposite sides of the fracture. When the

reservoir has the different permeability and porosity from each layer, the degree of

sweep and the length of the fracture should be different from each other. As a result,

the oil recovery is not as simple as for a homogeneous reservoir. Also, when the

reservoir properties are different for each layer, the fracture half length should be

different at each layer as well more accurate simulation could be conducted by

11

combining two simulators. Unlike the static fracture reservoir simulators, the

combination of two simulators could show the effect of the growing fracture with

different reservoir properties in each layer and direction.

After completing the combination of two simulators, the optimum fracture

length and orientation can be obtained. Also, the location of injectors and producers

will be considered. The variables that could be changed easily are the injection rate

and the concentration of solids in the injected water. The fracture growth rate is a

function of the injection rate, the temperature of the injected water and the

concentration of solids in the injected water. The fracture orientation will be changed

by altering the location of the producers. By changing the location of the producers,

the effect of the fracture orientation at each injection rate and concentration of the

injected water could be investigated. Similarly, the temperature of the injected water

will be changed and a sensitivity study for the temperature of the injected water will

be done. Finally, the optimum well pattern, spacing and injection rate will be

determined according to the reservoir properties.

A comparison will be made of vertical versus horizontal wells for different

reservoir geometries. Fracture growth in both types of wells will be studied to

determine when it may be appropriate to fracture these injection wells and when

fractures and producing wells would be desirable. The well spacing and well pattern

12

for maximizing reservoir sweep and injectivity will emerge from these simulation

results.

Task4: Simulation of EOR processes with fractured injectors and producers.

Surfactant and EOR process using surfactants and polymers will be simulated

to study the impact of fractures in injectors and producers. Both static and dynamic

fractures (fractures growing with time) will be studied. Vertical and horizontal wells

will be investigated to see how the time to first oil can be minimized and the reservoir

sweep maximized. It is expected that the formation of an oil bank in injection wells

and production wells are fractured will yield significantly different results than cases

where no fractures are present. The properties of the surfactant as well as polymer

will be varied so that the mobility ratio can be changed and the effect of the mobility

ratio can be clearly seen. To the best of our knowledge, this will be some of the first

systematic simulation of EOR processes using injectors and producers.

Task 5: Comparison of laboratory and field data

The results of the simulation from the combined model will be compared with

results obtained in EOR pilots conducted in the past. It is anticipated that these results

will primarily be for cases where no fractures are present. However, there have been

some recent reports of polymer injection field studies which have resulted in very

13

high injectivities presumably due to the creation of fractures. An attempt will be made

to compare the simulation results with the field results for both unfractured and

fractured injection wells. The advantage or disadvantage of using horizontal wells

with and without fractures will be studied for specific instances in the field where

EOR processes are proposed to be used.

14

4. Preliminary Results

Before combining the injection well simulator with a reservoir simulator,

some simple cases of water injection will run to investigate the role of fracture length

on reservoir sweep efficiency. All of these simulations were to investigate the impact

of static fractures with constant length on oil recovery.

Two patterns were chosen for the simulations. Figure 2-A. shows a fracture

oriented in an unfavorable direction so that the injected water migrated directly

towards the producers. Whereas Figure 2-B. shows a 5 spot pattern with a favorable

fracture orientation with the fracture propagating away from the producers.

Producer

Injector

Figure 2-A: Five spot injection pattern with unfavorable direction of fracture

15

Producer

Injector

Figure 2-B: Five spot injection pattern with favorable direction of fracture

Figure 3 shows the cumulative oil recovery normalized with the original oil in

place (OOIP) as a function of time for three different injection rates for an

unfractured well. The injection rate determines the oil production rate. However, the

cumulative oil recovery does not depend on injection rate. Figure 4 shows the same

results for three different well spacings for an unfractured well. The cumulative oil

recovery normalized with the original oil in place does not depend on the well

spacing. The time to reach the maximum production changes only as the well spacing

changes.

16

0.45

0.40

0.35
Oil Recovery / OOIP

"Injection Rate 1600 bbl / day"
0.30 "Injection Rate 800 bbl / day"
0.25 "Injection Rate 530 bbl / day"

0.20

0.15

0.10

0.05

0.00
0.0E+00 1.0E+04 2.0E+04 3.0E+04 4.0E+04 5.0E+04 6.0E+04
Time, days

Figure 3: Effect of injection rate on dimensionless production

0.050
0.045

0.040 Original Spacing
Oil Recovery / OOIP

0.035 Half Spacing
0.030 Double Spacing

0.025
0.020
0.015
0.010
0.005
0.000
0.0E+00 4.0E+02 8.0E+02 1.2E+03 1.6E+03 2.0E+03
Time, hours

Figure 4: Effect of well spacing on dimensionless production

17

Figure 5 and 6 show the effect of static fracture length and mobility ratio on

oil recovery. As the following figures show, the oil recovery is increased as mobility

ratio increases and as fracture length is decreased. The mobility ratio shows that the

mobility of the displacing material should be higher than that of the displaced

material to get the high oil recovery. This concept is used for increasing oil recovery

by polymer flooding as polymer flooding is used due to the high viscosity of

displacing material. Also, the reasons for lower oil recovery in a longer fracture are

early water breakthrough and poor sweep efficiency.

0.40

0.35
Oil Recovery / OOIP

0.30

0.25

0.20
mobility ratio=0.01
0.15 mobility=0.1
0.10 mobility ratio=1

0.05

0.00
0 0.2 0.4 0.6 0.8 1
Lf / r e

Figure 5: Effect of fracture length and mobility ratio on dimensionless production

18

0.255
mobility=0.1
Oil Recovery / OOIP 0.254

0.253

0.252

0.251

0.250

0.249

0.248

0.247
0 0.2 0.4 0.6 0.8 1
Lf / r e

Figure 6: Effect of fracture length on dimensionless production

19

5. References

Barkman, J.H. and Davidson, D.H.: “Measuring Water Quality and Predicting Well
Impairment”, J. Petrol. Tech., pp. 865 – 873, July 1972.

Gadde, P.B.: “Growing Injection Well Fractures and Their Impact on Waterflood
Performance”, SPE 71614 Presented at the 2001 SPE Annual Technical
Conference and Exhibition held in New Orleans, Louisiana, 20 September- 3
October 2001.

Pang, S. and Sharma, M.M.: “A Model for Predicting Injectivity Decline in Water-
Injection Wells”, SPE Formation Evaluation, pp. 194-201, September 1997.

Peaceman, D.W.: “Interpretation of Well-block Pressures in Numerical Reservoir
Simulation with Non-Square Gridblocks and Anisotropic Permeability” SPE
Journal, pp. 531-543, June1983.

Perkins, T.K. and Gonzalez, J.A.: “The Effect of Thermoelastic Stresses on Injection
Well Fraturing” SPE Journal, pp. 77-88, February 1985.

Saripalli, P.K., Bryant, S.L., and Sharma, M.M.: “Role of Fracture Face Plugging in
Injection Well Fracturing and Injectivity Decline”, SPE 52731 presented at the
1999 SPE/EPA Exploration and Production Environmental Conference held in
Austin, Texas, 28 February-3 March 1999.

Suarez-Rivera, R., Stenebraten, J., Gadde, P.B., Sharma, M.M.: “An Experimental
Investigation of Fracture Propagation during Water Injection”, SPE 73740
presented at the SPE International Symposium and Exhibition on Formation
Damage Control held in Lafayette, Louisiana, 20-21 February 2002.

Van den Hoek, P.J., Matsuura, T., de Kroon, M., and Gheissary, G. “Simulation of
Produced Water Re-Injection under Fracturing Conditions”, SPE 36846
presented at the SPE European Petroleum Conference held in Milan, Italy, 22-
24 October 1996.

Wang, D., Seright, R. S., Shao, Z., and Wang, J.: “Key Aspects of Project Design for
Polymer Flooding”, SPE 109682 presented at the 2007 SPE Annual Technical
Conference and Exhibition held in Anaheim, U.S.A., 11-14 November 2007.

20

Wang, G.C. and Caudle, B.H.: “Effects of Polymer Concentrations, Slug Size and
Permeability Stratification in Viscous Waterfloods”, SPE 2927 presented at the
1970 AIME Annual Fall Meeting in Houston, TX, U.S.A., 4-7 October 1970.

21

Lee - Effect of water injection above fracturing condition

More Related Content

What's hot (17)

Viewers also liked (13)

Similar to Lee - Effect of water injection above fracturing condition (20)

Lee - Effect of water injection above fracturing condition