UNIVERSITY OF OKLAHOMA GRADUATE COLLEGE ON THE GEOMECHANICAL OPTIMIZATION OF HYDRAULIC FRACTURING IN UNCONVENTIONAL SHALES A THESIS SUBMITTED TO THE GRADUATE FACULTY in partial fulfillment of the requirements for the Degree of MASTER OF SCIENCE By ALI TAGHICHIAN Norman, Oklahoma 2013
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
UNIVERSITY OF OKLAHOMA
GRADUATE COLLEGE
ON THE GEOMECHANICAL OPTIMIZATION OF HYDRAULIC FRACTURING
IN UNCONVENTIONAL SHALES
A THESIS
SUBMITTED TO THE GRADUATE FACULTY
in partial fulfillment of the requirements for the
Degree of
MASTER OF SCIENCE
By
ALI TAGHICHIAN Norman, Oklahoma
2013
ON THE GEOMECHANICAL OPTIMIZATION OF HYDRAULIC FRACTURING IN UNCONVENTIONAL SHALES
A THESIS APPROVED FOR THE MEWBOURNE SCHOOL OF PETROLEUM AND GEOLOGICAL ENGINEERING
BY
______________________________ Dr. Musharraf Zaman, Chair
______________________________ Dr. Ahmad Ghassemi
______________________________ Dr. Deepak Devegowda
First of all, I would like to express my deep gratitude and sincere thanks to my helpful and
kind advisor, Dr. Musharraf Zaman, because of all his help and support, key advices, and
his special style of encouragement; without these I would never be able to complete this
research work.
I am deeply grateful of my parents and my dear siblings for all their help, support, and
inspirations regarding my education at each level.
I also want to express my appreciation to Mr. Timothy L. Beard, Manager-ETG
operations, at Chesapeake Energy Corporation, Dr. Ali Daneshy, Head of Daneshy
Consultant International, for allocating their valuable time to me, advising me regarding
completion of unconventional resources, and being so patient and helpful.
I am indeed thankful of Dr. Arul Britto, Emeritus Faculty at University of Cambridge, Dr.
Jean-Claude Roegiers, Emeritus Faculty at University of Oklahoma, Dr. Ahmad Ghassemi
and Dr. Deepak Devegowda, faculty members of Mewbourne School of Petroleum and
Geological Engineering, University of Oklahoma for their kindness and sympathetic helps
and advices.
My gratitude also goes to Mr. Amin Mousavi, PhD candidate of Petroleum Geomechanics,
Tarbiat Modares University and Dr. Nicolas Roussel, Reservoir Engineer at ConocoPhillips
for their useful discussions.
Final thanks are also given to Oklahoma Department of Transportation (ODOT) and
Oklahoma Transportation Center because of their financial support during the course of
this study.
v
Table of Contents
Acknowledgements .......................................................................................................................... iv Table of Contents .............................................................................................................................. v List of Tables ................................................................................................................................... viii List of Figures ................................................................................................................................... ix Abstract .............................................................................................................................................. xi Chapter 1: Introduction .................................................................................................................... 1
1.1. Why unconventional resources ................................................................................... 1 1.2. Challenges and solutions in unconventional resources ........................................... 1 1.3. Critical questions in optimization of unconventional reservoirs ........................... 2 1.4. Approaches of fracing optimization in unconventional resources ........................ 3
1.4.1. Production approach ............................................................................................. 3 1.4.2. Geomechanics approach ....................................................................................... 3
1.5. Methods in geomechanical optimization of fracturing design ............................... 4 1.6. Strategies to control stress shadow and stress intensity factor ............................... 5 1.7. Contents of this thesis .................................................................................................. 6
Chapter 2: Literature Survey ............................................................................................................ 8 2.1. General ........................................................................................................................... 8 2.2. History of fracture mechanics ..................................................................................... 8 2.3. History of stress shadow around hydraulic fractures ............................................. 10 2.4. Methods of calculating stress field around cracks .................................................. 11
2.4.1. Stress field in the vicinity of crack tip ............................................................... 12 2.4.2. Stress field for the entire medium ...................................................................... 15
2.5. Calculation of stress shadow around hydraulic fractures ...................................... 17 2.5.1. Analytical determination of stress shadow size ............................................... 17 2.5.2. Numerical determination of stress shadow size .............................................. 18
2.6. Analytical calculation of crakc aperture ................................................................... 19 2.7. Need for a comprehensive study for optimization of fracing design .................. 20
Chapter 3: Methodology and Verification ................................................................................... 21 3.1. General ......................................................................................................................... 21 3.2. Significance of numerical methods for this problem ............................................. 21 3.3. Different scenarios for calculation of stress shadow size ..................................... 22 3.4. Major assumptions and data range in stress shadow analysis ............................... 24
3.4.1. Hydraulic fracture geometry ............................................................................... 25 3.4.2. Different boundary conditions for shadow analysis ....................................... 26 3.4.3. Assumptions ......................................................................................................... 26 3.4.4. Data range ............................................................................................................. 27
3.6.1. Parameters related to SIF change ...................................................................... 29 3.6.2. Monitoring of SIF change along the fracture edge by a single value ............ 31
3.7. Different scenarios for calculation of SIF change.................................................. 33 3.8. Data range for the SIF change analysis .................................................................... 35 3.9. Verification of stress field for simple problems ..................................................... 35
Chapter 4: Prediction of shadow and aperture of hydraulic fractures ..................................... 38 4.1. General ......................................................................................................................... 38
vi
4.2. Prediction of aperture for a contained hydraulic fracture ..................................... 38 4.2.1. Effect of in-situ stresses on aperture ................................................................. 38 4.2.2. Mathematical equation for prediction of aperture of a single contained
fracture ................................................................................................................... 39 4.2.3. Reliability of the proposed function for aperture prediction ......................... 40 4.2.4. Visualization of aperture prediction function .................................................. 41
4.3. Prediction of shadow around a contained hydraulic fracture (θT 5°) ............ 42 4.3.1. Numerical results for stress shadow analysis ................................................... 42 4.3.2. Mathematical equation predicting stress shadow of a single contained
fracture ................................................................................................................... 43 4.3.3. Coefficients of the proposed function for stress shadow prediction ........... 44 4.3.4. Reliability of the proposed function for stress shadow prediction ............... 46 4.3.5. Visualization of stress shadow prediction function ........................................ 47 4.3.6. Discussion on the relationship between aperture and stress shadow ........... 48
4.4. Prediction of shadow around a contained hydraulic fracture (θT 5°) ............ 48 4.4.1. Mathematical equation for prediction of stress shadow with different
threshold angles .................................................................................................... 49 4.4.2. Reliability of the equation predicting stress shadow with different threshold
angles ...................................................................................................................... 51 4.5. Effect of uncontainment of the fracture (standalone well fracturing) ................ 52
4.5.1. Change of stress shadow by uncontainment of the fracture ......................... 53 4.5.2. Change of aperture by uncontainment of the fracture ................................... 54 4.5.3. Mathematical equations predicting aperture and shadow size in uncontained
single fractures ...................................................................................................... 54 4.5.4. Reliability of the proposed functions for uncontainment multiplier
determination ........................................................................................................ 56 4.5.5. Use of uncontainment multipliers (standalone fractures) .............................. 57
4.6. Effect of simultaneous fracturing on shadow size and aperture .......................... 57 4.6.1. Change of stress shadow size by simultaneous fracturing ( 5°) .......... 57 4.6.2. Mathematical equation for prediction of simultaneous multiplier ................ 59 4.6.3. Reliability of the proposed equation for simultaneous multiplier prediction
60 4.6.4. Change of stress shadow size by simultaneous fracturing ( 80°) ........ 61
4.7. Effective distance between simultaneous fractures of two parallel wells ........... 61 4.8. Prediction of shadow change for uncontained simultaneous fractures .............. 62
4.8.1. Mathematical equations for uncontainment multiplier in simultaneous fracturing ............................................................................................................... 63
4.8.2. Reliability of uncontainment multiplier for simultaneous fracturing ............ 64 4.8.3. Use of uncontainment multipliers (simultaneous fractures) .......................... 64
Chapter 5: Prediction of Propagation potential in hydraulic fractures .................................... 66 5.1. General ......................................................................................................................... 66 5.2. Effect of aspect ratio of the fractures on the SIF (single fracture) ...................... 66 5.3. Effect of multistage fracturing on the SIF and aperture in standalone wells ..... 67
5.3.1. Qualitative description of multistage fracturing influence on SIF/aperture of each fracture ..................................................................................................... 68
5.3.2. Mathematical equation for prediction of SIF and aperture change in multistage fracturing ............................................................................................ 69
vii
5.3.3. Reliability of the proposed equation for SIF and aperture change in multistage fracturing ............................................................................................ 70
5.3.4. Visualization of the proposed equation ............................................................ 71 5.4. Effect of simultaneous fracing on the SIF of single-stage fractures ................... 72
5.4.1. Qualitative description of SIF change via simultaneous fracturing .............. 72 5.4.2. Mathematical equation for prediction of SIF change in simultaneous
fracturing (single stage) ........................................................................................ 73 5.4.3. Reliability of the proposed function .................................................................. 74
5.5. Effect of simultaneous multistage fracturing on the SIF of the fractures .......... 75 5.5.1. Qualitative description of the SIF change by simultaneous multistage
hydraulic fracturing .............................................................................................. 75 5.5.2. Mathematical equation for prediction of SIF change in multistage
simultaneous fracturing ....................................................................................... 77 5.5.3. Reliability of the proposed equation .................................................................. 78
5.6. Effect of fracture offset on the SIF change in parallel wells ................................ 79 5.7. Use of the SIF change prediction functions .................................................................. 82
Chapter 6: Work flow for optimization of hydrofracing ........................................................... 83 6.1. General ......................................................................................................................... 83 6.2. Successive procedure of hydraulic fracturing optimization .................................. 83
6.2.1. Step one ................................................................................................................. 83 6.2.2. Step two ................................................................................................................. 84 6.2.3. Step three ............................................................................................................... 85 6.2.4. Step four ................................................................................................................ 85
Chapter 7: Concluding remarks and recommendations ............................................................ 86 7.1. Aperture analysis ......................................................................................................... 86 7.2. Stress shadow analysis ................................................................................................ 87 7.3. Analysis of propagation potential ............................................................................. 87 7.4. Optimized fracture network ...................................................................................... 88 7.5. Four steps in optimization of hydraulic fracturing ................................................ 89
Thus, any increase/decrease in stress intensity factor directly results propagation to happen
with lower/higher hydraulic pressures, respectively. As a result, any change in stress
intensity factor is directly related to the amount of energy required for fracturing of the
formation and should be considered in different fracturing patterns as the concept to
examine fracturing potential change.
1.6. Strategies to control stress shadow and stress intensity factor
According to the fact that reducing stress shadow around hydraulic fractures without or at
least with the minimum decrease in propagation potential in the target zone brings the
opportunity to have more closely-spaced, non-deviating fractures, some technological
considerations have been reported in the literature in this regard. As stated previously,
simultaneous hydraulic fracturing is one of these methods performed in multi-lateral
horizontal wells. It is believed that the spacing between hydraulic fractures can be reduced
by utilizing this strategy. Mutalik and Gibson (2008) showed that simultaneously fractured
wells have 21-100% enhancement of initial production rates over the standalone horizontal
wells. King (2010) mentioned simultaneous fracturing as one effective strategy which
significantly raises the reservoir face-contact fractures. Waters et al. (2009) indicated
simultaneous fracturing as a powerful tool to reduce the spacing between fractures in
horizontal wells. Rafiee et al. (2012) studied simultaneous hydraulic fracturing of parallel
6
horizontal wells and modified the conventional approach of zipper frac to a more efficient
pattern called modified zipper frac. In this modified method, an alternating approach is
used for fracturing of parallel wells where far field path of hydraulic fracture is controlled
in a more acceptable way. Roussel and Sharma (2011b) suggested alternating fracturing
strategy to minimize the fracture spacing. In this method, stress shadow around outer
hydraulic fractures is used for making the middle fracture not to deviate. Of course, change
of propagation potential was not considered in their study.
Considering the most influencing variables in the size of stress shadow zone and stress
intensity factor of hydraulic fractures, aperture can also be predicted which is really useful
in term of proppant type/size and fracture conductivity determination.
Therefore, having the optimized distance between hydraulic fractures with no deviation or
collapse, having the propagation potential change in the acceptable range/direction, and
aperture of the fractures at hand, fracing crew can have an acceptable, optimized, and
desired fracturing design for an unconventional reservoir.
1.7. Contents of this thesis
In this thesis, it is tried to propose a work flow for optimization of hydraulic fracturing in
unconventional shales from a geomechanical point of view. Straight fractures without any
deviation/collapse, with the highest propagation potential in the target zone are the aim in
geomechanical optimization. This is done by utilizing two stress concepts called “stress
shadow” and “stress intensity factor”. In Chapter 2, the theoretical bases for these two
concepts are presented and the most important studies in these two fields regarding
optimization of hydraulic fracturing are also reported. In Chapter 3, the methodology of
7
determining stress shadow and stress intensity factor is presented. The most influencing
variables, the range of these variables, different boundary conditions, and different
mechanisms affecting the study are presented. Moreover, the need for investigating the
problem using numerical simulation is also reported and the used numerical technique is
verified by means of simulating simple crack problems and comparing the numerical with
analytical results. In Chapter 4, stress shadow and aperture of hydraulic fractures is studied
and two comprehensive sets of equations for estimation of stress shadow and aperture of
hydraulic fractures is proposed. In Chapter 5, stress intensity factor is investigated in a
comprehensive way and a complete set of equations for estimation of propagation
potential of hydraulic fractures having different geometries/patterns is proposed. In
Chapter 6, summing up the ideas regarding optimization of hydraulic fracturing in shales,
from production and geomechanics perspectives, a workflow is proposed for optimization
of hydraulic fracturing in shales. In Chapter 7, the summary and concluding remarks of
each step of the work is discussed and propositions are made in term of a systematic
method for optimization of hydraulic fracturing for this type of reservoirs.
8
Chapter 2: Literature Survey
2.1. General
In Chapter 1, the major factors in determining an optimized fracturing job in
unconventional reservoirs were introduced from a geomechanical point of view as stress
intensity factor and stress shadow effect. According to the fact that these two factors are
both indicative of stress field around hydraulic fractures, in this chapter, stress field around
hydraulic fractures is discussed in detail.
2.2. History of fracture mechanics
The birth point of obtaining stress field around cavities goes back to the work done by
Kirsch (1989), who succeeded to obtain stress field around a circular cavity in an infinite
two-dimensional (2-D) field. His closed form solution is still used for borehole stability
problems in elastic region. In this procedure, theory of elasticity together with complex
variable functions was utilized to obtain the stress function satisfying all the boundary
conditions. Then, Inglis (1913) obtained stress field around an elliptical cavity and proved
that by decreasing the curvature radius of the two ends of the ellipse, the increase in stress
concentration can be expressed by the following equation:
1 2 (2.1)
in which is the major axis of the ellipse and is the radius of curvature at the ellipse
sharp corners. As it is seen from Eq. (2.1), by tending the curvature of the corners to zero,
9
stress concentration approaches to infinity. This point was an eye-opening conclusion
which made engineers aware of the stress concentration at crack tips. Following this work,
a new field of mechanics was borne called “fracture mechanics”, in which stress field
around different fractures with different patterns and geometries were to be studied and
their propagation was to be pondered. Figure 2.1 shows an open fracture in a 2-D medium
with stress field around it.
Figure 2. 1. Configuration of stress field around a 2-D fracture
As it can be seen from Figure 2.1, a crack possessing two edges, have two half-lengths
called . The plate having this fracture inside can be applied by far-field tensile stress
( ;perpendicular to the crack plane) or internal pressure ( ). By having any of these two
stress/pressure boundary conditions; a stress field is created around the crack which can be
obtained.
r2
r1
r
X
Y
yy
xx
xy
yy
xx
P0
c c
10
2.3. History of stress shadow around hydraulic fractures
Many researchers mentioned the importance of stress shadow around hydraulic fractures.
Fisher et al. (2004) demonstrated that creation of a hydraulic fracture generates a zone of
altered local stresses that may impact the orientation of subsequent fractures in a
phenomenon known as the stress shadowing effect. Cheng (2009) utilized boundary
element method for 2-D geomechanical modeling of hydraulic fractures and indicated that
the number and spacing of the fractures need to be carefully selected considering stress
change in order to create effective fractures with appropriate geometries. Wong et al.
(2013) studied the interaction between adjacent hydraulic fractures using analytical and
numerical methods in two dimensions. They observed the divergence of hydraulic fractures
outward or even collapse of inside fractures on the outside ones as a result of stress
shadow effect. Singh and Miskimins (2010) indicated that an increase in spacing between
the fractures induced less interference, and hence requires less breakdown pressure to
initiate a fracture.
Waters et al. (2009) also stated that shadow following a hydraulic fracture treatment
involves the creation of a localized region of high compressive stresses perpendicular to the
fracture face in the vicinity of the fracture center. This causes the direction of maximum
stress to be reoriented in the region of the stress shadow. By locating the next treatment in
this region, fracture growth is likely to deviate or even occur parallel to the borehole axis
and consequently, necessitates optimizing fracture spacing to obtain the maximum number
of fractures oriented perpendicular to the wellbore (see also Roussel and Sharma, 2011a;
Morrill and Miskimins, 2012).
11
In general, closely spaced hydraulic fractures lead to an increase in the stress in the
minimum stress direction with values higher than the original maximum horizontal stress.
Succeeding fractures, therefore, may tend to propagate in undesirable directions parallel to
the wellbore axis, thereby reducing the efficacy of the fracture treatments and
compromising well productivity.
Stress shadow effect is a useful concept not only for determining the spacing between
transverse hydraulic fractures but also for fracture mapping and distance between
tectonically created fractures. Daneshy et al. (2012) were able to observe the shadow effect
from readings of pressure gauges installed in observation wells and used that information
for determination of fracture orientation and extent, spacing between wells, and optimum
number and spacing between fracture stages. Fischer et al. (1995) utilized the concept of
stress shadow for prediction of distance between pressurized fractures in sedimentary
rocks.
It is evident from the above-mentioned reviews that it is important to study stress shadow
of hydraulic fractures for optimization purpose.
2.4. Methods of calculating stress field around cracks
There are two different ways of determining the stress field around hydraulic fractures. In
the first method, it is assumed that radius of stress calculation point in polar system is
much smaller than the half-length of the fracture ( ≪ ; see Figure 2.1). Therefore, some
terms are neglected in the stress function, and a simple form of a complex variable
function is solved. In this method, stress field is obtained for only a small region around
the crack tip. According to the fact that propagation of a fracture is the main focus in
12
fracture mechanics and it mainly depends on stress field at the crack tip, this method
suffices for propagation potential of hydraulic fractures. In the second method, on the
other hand, no term is neglected from stress function and it is directly/numerically solved
for the entire field. This method is useful for determination of stress shadow around
hydraulic fractures.
2.4.1. Stress field in the vicinity of crack tip
Several researchers proposed the first solution for such problem as a 2-D crack embedded
in an infinite medium under far-field tensile stress or internal pressure. Among them,
works of Muskhelishvili (1933), Westergaard (1939), Sneddon and Elliot (1945), Paris
(1965), and Eshelby (1968) can be considered pioneering. They proposed that stress field
just around the crack tip can be obtained using Eq. (2.2), the notation of which
corresponds to the right tip ( , ) in Figure 2.1.
2cos
2
1 sin2sin
32
1 sin2sin
32
sin2cos
32
(2.2)
in which is the stress intensity factor (SIF) of the crack for mode one fracturing
(opening). It is worth indicating that both of our loading boundary conditions (far-field
tensile stress or internal pressure) belong to mode one loading. In addition, the SIF of a
crack is the same for the cases of far-field tensile stress and internal pressure with the same
13
magnitude (superposition principle; see Janssen et al., 2009). The generalized form of stress
field around the crack tip can be written as:
√2 (2.3)
It is important to notice that the trigonometric function, , is always the same for one
stress component. It is evident from Eqs. (2.2) and (2.3) that having different geometries of
cracks under mode one loading, only the SIF is changing and all the remaining terms are
kept unchanged. Therefore, having the SIF of each crack geometry, one can have stress
field around the crack tip. As an example, the SIF for this particular problem for far-field
stress is given by Eq. (2.4a) and for internal pressure it is calculated using Eq. (2.4b):
√2 (2.4a)
√2 (2.4b)
It is evident from Eqs. (2.4) that in case of equal magnitudes for far-field tensile stress and
internal pressure, the resultant SIF is the same. Moreover, it is seen that in the elastic
region, stress field is proportional to stress/pressure, square root of crack size, and
inversely related to the root distance from the tip.
Many 2-D crack geometries have been analytically solved and their SIFs have been
reported in the literature (e.g., Tada et al., 1973). However, there have also been some
problems in which the geometry of cracks were challenging and stress field for these
fractures were cumbersome to be analytically solved. SIF of such problems were defined by
14
utilizing Boundary Element (BE) and Finite Element (FE) methods (see e.g., Sih, 1973;
Tada et al., 1973; Murakami, 1987).
Regarding three-dimensional (3-D) crack problems, determination of stress distribution
near a flat crack, embedded in an infinitely extended homogeneous, isotropic solid, opened
up due to prescribed internal pressure, analytical solutions have been proposed by a
number of investigators for circular and elliptical geometries. Some of these solutions are
discussed by Keer (1964), Willis (1968), Sneddon and Lowengrub (1969), Shah and
Kobayashi (1971), and Guidera and Lardner (1975). Using the 2-D Fourier transform
method, Kassir (1981) succeeded in solving the SIFs around a rectangular crack. Moreover,
Mastrojannis et al. (1978) developed a method for determination of SIF of a general-
shaped crack with internal pressure in an infinite medium utilizing numerical integration.
By increasing of the internal pressure, SIF increases and reaches a critical value, called
fracture toughness, in which fracture tends to propagate and crack is elongated. It is worth
mentioning that different modes of fracturing have different fracture toughnesses.
For Mode I fracture toughness there exist ISRM (International Society of Rock Mechanics)
suggested methods to determine the fracture toughness (Ouchterlony, 1988; Fowell, 1995).
Several other methods have also been proposed in the literature (e.g., Evans, 1972; Barker,
1977; Atkinson et al., 1982; Chong and Kuruppu, 1984; Sun and Ouchterlony, 1986; Guo
et al., 1993; Chang et al., 2002, and Whittaker et al., 1992).
Based on the fact that SIF of fractures in three-dimensions is very challenging to be
analytically determined and in most of cases it is required to apply numerical integration,
15
therefore, researchers are using numerical techniques (e.g., FE and BE). As an example, in
hydraulic fracturing of unconventional shales with multistage simultaneous fracturing
strategy, it is important to consider the interaction of adjacent and meeting fractures on the
SIF. Numerical modeling is indeed a useful device in this regard because multi-driven
fractures in a medium may influence the SIF of each other and may change the
propagation potential. It is important to mention that propagation potential of a hydraulic
fracture is a relative term defined by the ratio of SIFs of multiple fractures over the SIF of
a single fracture. Thus, it gives a dimensionless number larger/smaller than unity because
of interacting crack tips effect.
2.4.2. Stress field for the entire medium
As explained in the previous section, Eqs. (2.2-2.4) only give stress field around the tip not
for the whole medium. This can be explained by considering a plate with a crack in the
center with far-field stress; if we put → ∞ in Eqs. (2.2), it is evident that all the stresses
tend to zero far away from the tip, which is not the case. Therefore, in order to judge the
disturbed stress zone or stress shadow size, a complete formulation for the whole field is
required to have all the components of stress at every point in the medium. In the world of
linear elastic fracture mechanics (LEFM), this is not an easy task to determine the stress
field for the whole medium, specifically for 3-D crack problems. Sneddon and Elliot (1945)
proposed Eq. (2.5) that describes stress field for the whole medium of a 2-D plate with an
internally pressurized crack in the center.
16
12
cos12
12
1 (2.5a)
12
sinsin
32
(2.5b)
sincos
32
(2.5c)
It is worth mentioning that the notation in Eqs. (2.5) corresponds to the notation in Figure
2.1. It is seen from Eqs. (2.5) that the stress field determined by six coordinate variables,
namely ( , ), ( , ), ( , ), is very challenging when such determination involves. The
stress field ( , , ) for this 2-D medium (e.g., E=30 GPa; 0.3) with an
imbedded center crack which is internally pressurized (e.g. 100 ) is presented as
Figure 2.2.
Figure 2. 2. Stress field around a center crack with internal pressure in a 2-D medium
It is seen from Figure 2.2 that stress concentration exists for all the components of stress at
crack tips. Shear stress is zero along the crack lips and on a line perpendicular to the crack
17
faces, passing through the crack center. This phenomenon happens because of the
symmetry in the model, which does not let the shear stress to exist on these symmetry
lines. The final point about Figure 2.2 is the contour shape of vertical stress ( ) around
the crack tips, which is a peanut-shaped inclined away from the crack tip. The plotted
contour shapes in Figure 2.2 exist for all the fracture geometries and can be used for
checking of the contour shapes in numerical simulations.
2.5. Calculation of stress shadow around hydraulic fractures
There are two different ways for determining the stress shadow size around hydraulic
fractures. In the first method, which is limited to some special cases of fracture geometries,
analytical stress field is used for determination of shadow size. In the second method,
numerical simulation is used for determination of stress field around hydraulic fractures
with any geometry and boundary conditions. Then, stress shadow is calculated from the
stress field. In the following sections both of these methods are presented.
2.5.1. Analytical determination of stress shadow size
Jo (2012) proposed an analytical method for prediction of stress shadow based on
analytical stress distribution around plane strain and penny-shaped fractures (see Green
and Sneddon, 1950). This method can only be used for plane strain and penny-shaped
cracks. Equations (2.6) show the proposed equations for predicting shadow size around
these two limiting cases of hydraulic fractures.
2 3 2
for plane strain fractures infinite length
14 5
for penny shaped fractures AR unity
(2.6)
18
in which, is the distance from the center of a hydraulic fracture to a point where
maximum stress contrast between and is observed, is the Poisson’s ratio, and
is height of the hydraulic fracture. From Eq. (2.6) it is seen that the only influencing
variables in the shadow size are the Poisson’s ratio and fracture height. In addition, Eq.
(2.6) accounts for the aspect ratio but it only addresses stress shadow size for two limiting
cases of plane strain and penny-shaped cracks. Effect of internal pressure, stress
anisotropy, and aspect ratios of the hydraulic fracture, however, have also been found to
influence stress shadow size (Roussel and Sharma, 2011a,b; Morrill and Miskimins, 2012).
2.5.2. Numerical determination of stress shadow size
In this method, having the total stress field from numerical simulation, the change in
magnitude and direction of the principal stresses can be obtained from Eq. (2.7) and Eq.
(2.8), respectively.
, 2 2
(2.7)
tan 22
(2.8)
in which , , are induced stresses in the shadow region, and are changed
principal stresses in this region, and is the direction of principal stresses. Deviation
angle of principal stresses ( ) can also be obtained based on the following criteria:
19
9045
(2.9)
It is worth mentioning that maximum principal stress deviation, depending on the
magnitude of internal pressure, can be near the crack tip or crack center and it vanishes
away from the crack. Therefore, by defining a threshold angle ( ) the area around the
crack with stress deviations more than this threshold angle can be determined. Considering
Eq.(2.8), it is also evident that maximizing the stress contrast means lowering the deviation
angle. Using this numerical method, there is no limitation on the fracture geometry and
boundary condition. Thus, a wide range of problems can be solved satisfactorily using this
approach.
2.6. Analytical calculation of crakc aperture
Sneddon and Elliot (1946) also documented the analytical solution for crack opening
displacement field. This analytic expression for half-aperture of a 2-D hydraulic fracture
was derived as:
1 ;2 1
(2.10)
in which is the Poisson’s ratio, is the Young’s modulus of net play, is the
hydraulic pressure applied inside the fracture, is the half length of the fracture, is
the half aperture, is the maximum half-aperture (located at the center of the fracture
( 0), and is defined on the z axis originated at crack center in the direction of fracture
20
height. From Eq. (2.10), we observe that the displacement of crack tips (at ) is zero
and displacement of the edges increases from the tips to the center of the fracture where
maximum displacement occurs. The equation also highlights the inverse relationship of
fracture aperture with Young’s modulus and Poisson’s ratio of the rock. Furthermore, the
equation describes an elliptical shape for the aperture of the fracture. It is important to
note that for a 3-D hydraulic fracture, maximum aperture is located at half-height and half-
length of it. So, an equation similar to Eq. (2.9) is written in direction, along the length of
the hydraulic fracture. Therefore, having the maximum aperture of the fracture, one can
use Eq. (2.9) and a similar one for direction for determination of aperture distribution on
the entire crack face.
2.7. Need for a comprehensive study for optimization of fracing design
According to the above-mentioned studies in the field of stress shadow and propagation
potential of hydraulic fractures, it is realized that most of these studies have been
performed in two dimensions or are more or less descriptive. Therefore, it can be seen that
there is a lack of comprehensive 3-D studies of stress shadow with propagation potential.
In this thesis, we plan to comprehensively study the stress shadow size and propagation
potential of hydraulic fractures together. In the following chapter, the methodology of
solving the present problem together with verification of the method is presented.
21
Chapter 3: Methodology and Verification
3.1. General
In Chapter 2, basic concepts of stress distribution around hydraulic fractures with
particular attention to propagation potential and stress shadow effect in the target medium
were explained using a 2-D infinite medium with an internally pressurized fracture in the
middle. In addition, a clear perspective from which optimization of hydraulic fracture
should be seen, was introduced by explaining the two concepts of stress intensity factor
(SIF) and stress shadow effect. In this chapter, the method by which we are going to
investigate the problem of optimization of hydraulic fractures in unconventional shale
reservoirs is introduced. The presented method is verified by solving a number of single
fractures in 2-D/3-D media with simple geometries and making a comparison between the
obtained results and analytical solutions reported in the literature for these problems.
3.2. Significance of numerical methods for this problem
Obtaining the stress field around a fracture with its own geometry and boundary
conditions, in a 3-D medium is a challenging matter. Although there are some analytical
solutions available in the literature for some cracks with simple geometries, but finally, a
numerical integral should be solved to obtain the stress field for these problems. As a
result, because of the limited flexibility in analytical solutions, a numerical method is
utilized in this thesis to study the problem of optimization of hydraulic fractures. Finite
Element Method (FEM) which is a powerful tool in solving governing equations of
problems in solid/fluid or coupled solid-fluid mechanics, is chosen for this purpose.
22
ABAQUS CAE 6.12, which is a general purpose FEM-based software package, is selected
for the modeling purposes. The procedure of modeling is designed in such a way to first
give us the opportunity to obtain stress shadow size which is indicative of hydraulic
fracture distance, based on different influencing variables. Then, propagation potential of
these hydraulic fractures is studied in detail to examine which geometry/pattern has what
effect on the propagation potential of the fractures. Finally, according to the constructed
numerical models, aperture of the hydraulic fractures which is highly important in the
selection of proppant type and size and determination of fracture conductivity is estimated
and its change with the defined variables is studied.
3.3. Different scenarios for calculation of stress shadow size
Four different scenarios are defined in this section, based on which stress shadow and
aperture analysis of hydraulic fractures are performed (Figure 3.1).
Figure 3. 1 Different scenarios for studying of stress shadow and aperture (pictures show fracture plane)
SingleContainedHFstandalonewell
SimultaneousContainedHFParallelwells
SingleUn‐containedHFstandalonewell
SimultaneousUn‐containedHFParallelwells3
1 2
4
23
In the first scenario (No. 1 in Figure 3.1), a 3-D hydraulic fracture, with different aspect
ratios is contained in a medium with different stress regimes, moduli, and net pressures. In
Chapter 2, theoretical background of stress shadow was presented and it was shown that
assuming different threshold angles, different shadow sizes are obtained (see Chapter 2,
Section 2.5.1). According to the fact that in the literature (see Roussel and Sharma, 2011),
region of 90° reorientation of maximum horizontal stress is assumed as the region of
minimum shadow effect, hence, it is tried to predict the shadow with all the threshold
angles.
In the second scenario (No. 2 in Figure 3.1), it is assumed that the hydraulic fracture under
consideration is not contained. Since shale net plays are located in between bounding layers
with different moduli from the net play, it is important to consider any fracture penetration
inside the bounding layers. This is done by assuming different Young’s modulus between
the net play and bounding layers, and different penetration extent to the bounding layers.
In the third and fourth scenarios (Nos. 3 and 4 in Figure 3.1), stress shadow change as a
result of simultaneous hydraulic fracturing of laterals is studied in detail. Effect of distance
between hydraulic fracture tips in parallel wells is investigated on the shadow size. In this
part of the work, effective length between fracture tips is also obtained for fractures with
different aspect ratios. It is worth mentioning that effective distance is a distance between
fracture tips beyond which no shadow size change is observed. In addition, effect of
penetration of the simultaneous hydraulic fractures inside the bounding layers (uncontained
fractures) is studied and the shadow change is investigated.
24
Using the above mentioned scenarios, stress shadow and aperture of a hydraulic fracture
can be predicted satisfactorily. First, a set of equations is proposed for a single contained
hydraulic fracture (Scenario 1). After that, in Scenario 2, effect of bounding layers for the
hydraulic fracture is presented as a ratio to be multiplied with the result of the first scenario
to result the value for an uncontained hydraulic fracture. Likewise, in the third scenario, a
ratio is obtained which is the effect of simultaneous fracturing for contained fractures. This
ratio should also be multiplied with the result of the first scenario to provide the result for
the case of simultaneous contained fractures. Finally, for the fourth scenario, effect of
uncontainment of simultaneous fractures is obtained as a ratio to be multiplied with the
result of the third scenario to give the results for simultaneous uncontained fractures.
As an example, in case one is interested in obtaining the shadow size of uncontained
simultaneous fractures between two parallel wells, it is first required to have the shadow for
the basic case; contained hydraulic fracture in a standalone well (Scenario 1). Then, effect
of simultaneous fracturing of parallel laterals is obtained for a contained fracture (Scenario
3) as a ratio to be multiplied with the result of the basic case. Finally, effect of
uncontainment of simultaneous fractures (Scenario 4) is obtained as a ratio to be multiplied
with the resultant value in the previous step.
3.4. Major assumptions and data range in stress shadow analysis
Geometry of the hydraulic fracture with respect to the net play and bounding layers,
different boundary conditions, the most influencing variables and their varying range, and
assumptions of the analyses are defined in this section.
25
3.4.1. Hydraulic fracture geometry
Geometry of the hydraulic fracture with respect to the wellbore, net play, and bounding
layers is shown in Figure 3.2a. As it can be seen from this figure, a rectangular hydraulic
fracture located in plane is made perpendicular to the wellbore. According to the fact
that maximum stress change is observed in the center of the hydraulic fracture (see Waters
et al., 2009), a plane perpendicular to the height axis ( direction) passing through the half-
height of the fracture is used for stress shadow, and maximum aperture analysis. Q plane in
Figure 3.2b shows the plane of aperture and stress shadow calculation. Length, height, and
width (aperture) of the hydraulic fracture are also shown in Figure 3.2b. It is necessary to
indicate that in this analysis, (half of the maximum aperture; see Eq. (2.10)) is
numerically calculated and can be predicted using the proposed equations in the following
sections. Therefore, aperture of the hydraulic fracture can simply be obtained by doubling
the value of .
Figure 3. 2. Geometry of a contained hydraulic fracture and stress shadow plane
Boundinglayer
wellbore
c
z
yx
H
width
Length:2c
Height:H
Aperture:2Wmax
Boundinglayer
NetPlay
Q
(a) (b)
26
3.4.2. Different boundary conditions for shadow analysis
In general, modeling approaches for evaluating the stress shadow effect adopt either one of
two boundary conditions for their case studies. These are:
1. Choice of a fixed hydraulic pressure inside the fracture ( ): This implies that
the fracture aperture is changed by varying values of the moduli of the rock and
hydraulic pressure inside the fracture.
2. Choice of a fixed maximum fracture aperture ( ): With a fixed fracture
aperture, the influence of other parameters on the stress shadowing may be
investigated.
The assumption of a constant pressure within the fracture mentioned as the first boundary
condition is a more reasonable approximation of reality and therefore, in this study, we
employ this boundary condition.
3.4.3. Assumptions
The key assumptions employed in the numerical simulation are:
The 3-D domain is assumed to be a completely elastic medium without any
plasticity-based constitutive laws;
Propagation of the fracture is not considered in this work; instead the hydraulic
fracture is assumed to have been created;
27
The aspect ratio defined as the ratio of height to the length of the fracture is
equal/less than unity;
The threshold angle ( ) is varied from 5°-80°;
The aperture of the fracture is 2 and therefore is twice the ;
In case shadow analyses are carried out for a threshold angle of 80° ( _
and _ ), only analyses with hydraulic pressures larger than maximum
horizontal stress ( ) are considered (see shadow mechanism part;
Section 3.5).
3.4.4. Data range
In this section, the adopted data range for the massive numerical simulation is presented.
Since stress shadowing is most challenging when we have lower stress anisotropies (see
Morrill and Miskimins, 2012), the ratio of minimum horizontal stress over maximum
horizontal stress is assumed to be in the range of 0.95-0.99. Moreover, previous
studies in shale have indicated that Young’s modulus of shale changes from 7 to 77 GPa
with an average value of 26.9 GPa (Hay and Sondergeld, 2011). Therefore, Young’s
modulus of the net play is assumed in the range of 10-70 GPa. For studying of the
effect of bounding layers, modulus of bounding layers is assumed as 0.25-4.0 times of that
of net play and 0.1-0.3 of the fracture height is assumed to penetrate inside the bounding
layers. For the case of simultaneous hydraulic fracturing of horizontal wells, the distance
between parallel wells is varied with similar fracture length and aspect ratios for both of the
wells. The input variables and their ranges are shown in Table 3.1.
28
Table 3. 1. Input variable range for the numerical simulation Parameter Range Unit Fracture aspect ratio 1.00-0.20 - In-situ stress anisotropy 0.95-0.99 - Poisson’s ratio 0.00-0.40 - Young’s Modulus 10-70 GPa Excess pressure 0.0-3.0 MPa Young’s modulus ratio1 0.25-4.0 - Penetration2 0.1H-0.3H m Distance between tips 3c -
3.5. Shadow mechanisms
In order to study the stress shadow, different mechanisms of shadow around a hydraulic
fracture should be recognized. The mechanisms of stress shadowing generally belong to
two categories which will be investigated in this study:
1. The pressure applied lies between the minimum and maximum horizontal
stress ( ). This results in only a marginal deviation of
the in-situ stresses around the crack tips ( 45°) and consequently,
does not lead to a principal stress reversal (90° rotation). Figure 3.3a
illustrates a marginal deviation of 40° for this scenario.
2. In cases where the applied pressure is larger than the maximum horizontal
stress ( ), a reversal of stress occurs ( 45°) and this
phenomenon is illustrated in Figure 3.3b.
It should be noted that principal stress deviations plotted in Figs. 3.3 are from the
maximum deviation plane (plane Q defined in Figure 3.3b). The pink lines show the
fracture plane.
1 Young’s modulus ratio is the ratio of Young’s modulus of the bounding layer over that of the net play 2 Penetration is the part of the fracture height penetrating the bounding layers
Figure 3.
Consequentl
fracture and
mechanism
shadow. It i
only be ana
zero for thre
3.6. Met
In this secti
edge. In thi
studied alon
3.6.1. Param
According t
knowing the
3. Different
ly, stress sha
may cause a
difference s
is important
lyzed for th
eshold angles
hod of stud
on, the tech
is thesis, as
g length and
meters related t
o Irwin (195
e SIF of the c
t stress shad
adow is seen
a diverse ran
should be k
to notice th
e second me
s of larger tha
ying stress i
nique for inv
a result of
height of th
to SIF change
57), stress fie
crack. Consid
(a)
29
dow mechan
n to be a fun
ge of deviati
ept in mind
hat stress sh
echanism be
an 45°.
intensity fac
vestigating o
assuming re
e fractures.
eld around th
dering Eq. (2
Dev
iatio
n an
gle,
nisms by dif
nction of the
ions in the p
d when inter
hadow for th
ecause shado
ctor
of SIF chang
ectangular fr
he crack tip
2.2), we have:
fferent hydra
e hydraulic p
principal stres
rpreting the
hreshold angl
ow for the fi
ge is present
ractures, SIF
can be dete
:
aulic pressu
pressure insid
ss directions.
results of
les of 80° sh
irst mechani
ed for each
F change is t
rmined simp
(b)
ures
de the
. This
stress
hould
ism is
crack
to be
ply by
Dev
iatio
n an
gle,
30
∝ (3.1)
in which is linearly related to the internal pressure ( ), directly related to a function of
the geometry of the fracture ( ), and a function incorporating the effect of boundaries on
the fracture ( ). Effect of boundary represents itself as the effect of having cracked media
with finite dimensions, or having multiple adjacent fractures. Therefore, we can write:
(3.2)
As an example, for a 2-D fracture in an infinite medium under internal pressure (see
Chapter 2), √ and 1. Determining , for different crack geometries and
different boundary conditions is of paramount importance. This is because by having these
functions, one can determine the SIF of a specific crack geometry, thereby having the
stress field around the crack tip and consequently, its potential for propagation.
Combining Eqs. (3.1) and (3.2), we can write:
∝ (3.3)
Therefore, ratio of a stress component over the internal pressure can be assumed as
indicative of the effect of crack geometry and its boundary conditions on the SIF.
According to the fact that in this thesis, we intend to change the geometry of the hydraulic
fracture (aspect ratio) and consider different fracture patterns (simultaneous and multistage
fracturing), all the changes in the SIF can be considered as a result of changes in these two
functions. Therefore, by the above-mentioned changes, the ratio of stress over the internal
31
pressure will result the SIF change as a result of geometry and boundary condition
functions ( .
3.6.2. Monitoring of SIF change along the fracture edge by a single value
Crack tip of a 3-D hydraulic fracture is a closed line in space which may be an irregular
shape depending on many parameters including in-situ stress, hydraulic pressure change,
moduli variation, containment, heterogeneity of the medium, etc. In this study, this closed
loop has been simplified to a rectangle along which the SIF is changing by position (Figure
3.4).
Figure 3. 4. Fracture geometry together with stress change along its edges
Therefore, according to the prescribed mesh size of the FE model, a large number of data
(stress components) are associated with each case of study. In order to make a comparison
much simpler between different geometries/fracturing patterns, a function is fit to the data
and the resultant coefficients of the function are compared instead. The proposed function
which gives satisfactory fit on the numerical stress values is as given in Eq. (3.4).
Length,2c
Height,2b
x
y
zz
zz
32
1 (3.4a)
1 (3.4b)
in which is the applied hydraulic pressure, is the normal stress at the corner, is
normal stress along the length or height of the fracture, axis is in the direction of length
originated from the corner, is the fracture half-length, is the axis in the direction of
fracture height originated from the corner, is the half-height of the fracture, and are
coefficients of the function ( 1,2). Investigating the proposed function, it is revealed
that stress change is highly controlled by rather than . This is because the negative
sign of in an exponential format makes a small value to be much smaller. Since this
exponential value is subtracted from unity, no noticeable change is observed for stress in
the coefficient range we are dealing with in this regard.
Figure 3.5 shows a typical example of normal stress change along the crack edge. As shown
in the figure, behavior of stress change along crack length or height has satisfactorily been
predicted using Eqs. (3.4). It is also evident from the figure that normal stress is maximum
in the middle point of the crack length and it is reduced to the minimum value at the crack
corner. Since normal stress is normalized by being subtracted by stress at the corner and
divided by the internal pressure, the values at the corners are zero. In addition, stress
change with height or length of fractures with aspect ratio of unity is similar as expected
but for cracks with aspect ratio lower than unity, stress decreases with both length and
height of the fracture, however, the magnitude of decrease is higher in fracture height
33
compared to that of the length. The conspicuous point is that despite the fact that length
of the fracture is not changing and aspect ratio is being reduced by decreasing the fracture
height, still stress along the crack length is reduced.
Figure 3. 5. Stress variation along the fracture edges
Considering the dominant coefficient of the function proposed in Eqs. (3.4), one can have
a good comparison between SIF of different crack geometries and multi-fracture
configurations. As it can be seen from Eqs. (3.4), this value is multiplied with the
exponential function and as a result, any difference in this coefficient means the same
change in normal stress and consequently, the SIF of the crack.
3.7. Different scenarios for calculation of SIF change
There are four different scenarios that are considered for SIF change analysis in this thesis.
Figure 3.6 shows the adopted scenarios for the hydraulic fracturing pattern/technique.
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Norm
alized norm
al stress, ( z
z‐ z
z 0)/PH
Normalized tip position, (x/c,y/b)
Szz_Height (AR=1.0)
Szz_Length (AR=1.0)
Szz_Height (AR=0.2)
Szz_Length (AR=0.2)
1.00 0.0 1.00
34
Figure 3. 6. Different fracturing techniques/patterns in this study
It is evident from Figure 3.6 that Scenario 1 is the basic scenario in which we have a single
stage fracture from a standalone well. Scenario 2 shows multistage hydraulic fracturing in a
standalone well. In order to compare propagation potential of the second scenario
(multistage fracturing) with the basic case (Scenario 1), the numerically obtained SIF of the
second scenario is divided by the SIF of the first scenario. Any value lower/higher than
unity means that propagation potential for the scenario under study is smaller/larger than
that in the basic case. Scenario 3 considers the effect of simultaneous single stage fracturing
of the medium between two parallel wells. The obtained SIF for this scenario should also
be compared with that in scenario one in order to quantify the effect of simultaneous
fracturing in a single stage mode. Finally, Scenario 4 is studied in which the effect of
simultaneous multistage hydraulic fracturing is considered between two parallel wells.
Scenario 4 has two influencing variables acting simultaneously on the SIF of the fractures.
The resultant SIF from this scenario should be compared with results of Scenario 1 in
order to have the effect of simultaneous multistage fracturing pattern on propagation
potential of the fractures.
1 2
3 4Ls
2Lp
35
Therefore, changing the spacing between fractures in multistage fracturing and distance
between the meeting fractures in simultaneous fracturing brings the opportunity of
quantifying the SIF change for each scenario. Of course, the effect of fracture geometry
(aspect ratio) on the SIF is also considered.
The above-mentioned scenarios are established to quantify the SIF change according to the
geometry and pattern of the fractures. One qualitative analysis is also done on the effect of
offset between meeting fractures on the SIF change of the fractures. The term offset in this
regard means the distance between two planes on which hydraulic fractures are located in a
simultaneous fracturing mode.
3.8. Data range for the SIF change analysis
In this study, an effective range has been assumed for all the studies. The input variables
and their range are defined in Table 3.2.
Table 3. 2. Input variable range for the numerical simulation No. Parameter Range1 Fracture aspect ratio 0.2-1.02 Fracture spacing1 0.1 -7.0 3 Fracture distance2 0.1 -3.0 4 Fracture offset 0.1 -1.0
Table 3.2 shows that an extensive number of numerical models are required to be built and
analyzed to predict the SIF change via simultaneous multistage fracing job.
3.9. Verification of stress field for simple problems
In this section, the modeling strategy used in ABAQUS is examined by numerically solving
simple crack geometries, the analytical solutions of which are available in the literature. 1 The term “spacing” in this thesis is only used as a measure of the interval between the adjacent fractures in multistage fracturing mode. 2 The term “distance” in this thesis only refers to the distance between meeting fracture tips in simultaneous fracturing mode.
36
Semi-infinite and penny-shaped cracks with internal pressure are chosen for this purpose
(Sneddon, 1945). In this step of the work, appropriate size of the model is defined in order
to minimize the boundary effects to the model response.
Cracks in a material are associated with three geometrical parameters; length, width (called
aperture), and height. The ratio of any two of these can be considered as the aspect ratio.
In this paper, the ratio of height over length of the fracture is called aspect ratio (
/ ). We numerically simulate two cracks with different geometries, one with
an infinite height (plane strain crack; ∞), and the other with aspect ratio of unity
(penny-shaped crack; 1.0), both with an internal pressure. The results of these
numerical simulations are compared with the analytical solutions.
The validation approach described herein compares the horizontal and vertical stresses
around the simulated crack with the analytical results presented by Sneddon and Elliot
(1946). These are shown in Figure 3.7 for horizontal and vertical stresses along a line
perpendicular to the face of the hydraulic fracture as shown in the inset.
Figure 3. 7. Stress validation in the direction vertical away from the fracture center
‐1.20
‐1.00
‐0.80
‐0.60
‐0.40
‐0.20
0.00
0.20
0 2 4 6 8 10
Norm
alized
Stress, S/P
H
Normalized distance, x/c
plane strain crack
SXX/PH (NUMERICAL)
SYY/PH (NUMERICAL)
SXX/PH (ANALYTICAL)
SYY/PH (ANALYTICAL)
xx
xx
yyyy
‐1.2
‐1
‐0.8
‐0.6
‐0.4
‐0.2
0
0.2
0 1 2 3 4 5
Norm
alized Stress, S /PH
Normalized distance, x/c
Penny‐shaped crack
Syy/P (ANALYTICAL)
Sxx/P (ANALYTICAL)
Syy/PH (NUMERICAL)
Sxx/PH (NUMERICAL)
xx
xx
yyyy
37
The figure demonstrates excellent agreement between the analytic and numeric simulation
results for plane-strain and penny-shaped cracks and, therefore, the numerical modeling
strategy adopted here can be considered valid for modeling of hydraulic fractures and stress
shadowing effects. As a result, according to the used numerical method and all the
conditions/assumptions we can run numerical simulations to study the influence of the
considered variables on stress shadow size and stress intensity factor. In Chapters 4 and 5,
all the results of numerical simulations are presented.
38
Chapter 4: Prediction of shadow and aperture of hydraulic fractures
4.1. General
In Chapter 3, the methodology of investigating stress shadow and stress intensity factor
(SIF) of a hydraulic fracture was proposed. All the applied boundary conditions and major
assumptions were also presented. In this chapter, based on explanations in Chapter 3, the
results of all the analyses conduced for stress shadow of hydraulic fractures are reported
and a detailed discussion on the stress shadow is also presented.
4.2. Prediction of aperture for a contained hydraulic fracture
First, aperture of hydraulic fractures is measured according to the numerical results and it is
predicted by proposing an equation. This is done first because aperture of hydraulic
fractures is of crucial significance in proppant size/type and estimating fracture
conductivity.
4.2.1. Effect of in-situ stresses on aperture
As mentioned in Chapter 2, Eq. (2.10) describes half-aperture of a hydraulic fracture in a 2-
D medium as a function of the applied hydraulic pressure, relative distance from fracture
center, and rock moduli. The formulation of Eq. (2.10) ignores the impact of any far-field
stress. However, for the case of aperture determination, when in-situ stresses are present,
the excess hydraulic pressure ( ) should be considered instead. This is
because according to the numerical results, maximum horizontal stress (applied parallel to
the fracture face) has negligible effect on the aperture of the internally pressurized fracture
(see also Jeffrey, 1989) and therefore, a simple subtraction of minimum horizontal stress
from the applied pressure and neglecting far-field stresses suffices to estimate the aperture
39
similar to the case mentioned in Eq. (2.10). Of course, this elimination is only valid for the
case of aperture, where stress anisotropy is not an influencing variable.
4.2.2. Mathematical equation for prediction of aperture of a single contained fracture
In this section, aperture of a contained hydraulic fracture in a standalone well is predicted
considering different mechanical properties for the shale, variant stress anisotropies,
different internal pressures, and fracture aspect ratios (see Scenario 1, Figure 3.1; Chapter
3).
First, it was observed that there is similar effect of Young’s modulus in aperture
determination of 2-D and 3-D fractures. Influence of Poisson’s ratio, on the other hand,
was observed to be different from the 2-D fractures in such a way that aperture of a 2-D
fracture shows a quadratic dependence on Poisson’s ratio (see Eq. (2.10)), while for a 3-D
hydraulic fracture, there is a third order polynomial dependence between aperture and
Poisson’s ratio. In addition, aperture of 3-D hydraulic fractures shows a quadratic
dependence on the fracture aspect ratio as well. Following the above explanations, aperture
of a contained, 3-D hydraulic fracture can be determined using Eq. (4.1).
, , (4.1)
in which is a third order polynomial of the Poisson’s ratio, multiplier is a
quadratic function of aspect ratio, is maximum half-aperture, and , , are
excess pressure, net play Young’s modulus, and half length of the hydraulic fracture,
respectively. These functions are defined in Eqs. (4.2) and (4.3):
40
;1.1286 0.89190.1494 2.0286 (4.2)
0.2757 1.3092 0.0316 (4.3)
This equation underscores the independence between the stress anisotropy and the half-
aperture of hydraulic fractures, implying that aperture only depends on the minimum
horizontal stress ( ).
4.2.3. Reliability of the proposed function for aperture prediction
A comparison of the numerical results and the predictions for Eq. (4.1) are shown in
Figure 4.1 which demonstrates excellent agreement between numerically obtained values
and the predicted ones.
Figure 4. 1. Half-aperture prediction using Eqs. (4.1-4.3)
Therefore, employing Eqs. (4.1-4.3), one can have a satisfactory estimate of the aperture of
a hydraulic fracture by giving the aspect ratio, Poisson’s ratio, excess pressure, and Young’s
modulus. Excess pressure of a hydraulic fracture is determined via subtraction of minimum
horizontal stress from hydraulic pressure inside the fracture. Therefore, minimum
w'max = 0.9991 wmax
R² = 0.9999
0.00E+00
1.00E‐04
2.00E‐04
3.00E‐04
4.00E‐04
0.00E+00 1.00E‐04 2.00E‐04 3.00E‐04 4.00E‐04
Predicted half‐ap
erture ratio, w
' max/ c
Numerical half‐aperture ratio, wmax / c
41
horizontal stress should also be considered as an influencing variable in determination of
aperture of hydraulic fractures.
4.2.4. Visualization of aperture prediction function
The proposed equation for prediction of aperture in 3-D fractures has four inputs
(Poisson’s ratio, aspect ratio, Young’s modulus, and excess pressure) and aperture as the
function of these inputs. Visualization of this function is only possible when plotting is
done by the ratios, and having different plots for different aspect ratios.
The expression for the fracture half-aperture in Eq. (4.1) is re-drawn in Figure 4.2 for
different Poisson’s ratios, excess hydraulic pressures, two aspect ratios, and two different
values of Young’s moduli.
Figure 4. 2. Visualization of aperture determination function
The key observation from Figure 4.2 is the relationship between Young’s modulus and
fracture half-aperture. More compliant rocks characterized by a lower Young’s modulus are
associated with a larger fracture aperture. Increases in excess pressure also lead to a
widening of the fracture and so do decreases in the value of Poisson’s ratio. In order to
42
justify the behavior of aperture with respect to moduli of the rock, consider elastic moduli
relationship. Bulk compressibility of rock is obtained as 3 1 2 / . This means
by decreasing of Poisson’s ratio and Young’s modulus, compressibility of rock increases,
thereby raising the aperture of the hydraulic fracture.
4.3. Prediction of shadow around a contained hydraulic fracture ( °)
This section focuses on the description of the stress shadowing around a single contained
hydraulic fracture (see Figure 3.1; Scenario 1 in Chapter 3). The input parameters were
varied according to Table 3.1 to quantify the shadow size around a hydraulic fracture. It is
worth mentioning that because of using pressure boundary condition, Young’s modulus of
the medium does not show any effect on stress shadow.
4.3.1. Numerical results for stress shadow analysis
In order to study the stress shadow, excess pressure ( ) is normalized with ten times of
the atmospheric pressure (10 ) and shadow length is normalized with the fracture half
length ( ). The normalized shadow lengths predicted as a function of the normalized
excess pressures are typically shown in Figure 4.3 for an aspect ratio of 1.0, horizontal
stress anisotropy of 0.95 and varying Poisson’s ratios.
It can be seen that there is a nonlinear direct relationship between the shadow size and
internal pressure. This means that by raising the internal pressure, shadow size increases
but with a decreasing gradient. In Figure 4.3, we also describe the variations in the stress
shadow effect with Poisson’s ratio which shows that shadow size increases with Poisson’s
ratio.
43
Figure 4. 3. Shadow change with Poisson’s ratio, numerical values and the predictions
4.3.2. Mathematical equation predicting stress shadow of a single contained fracture
We use all the numerical results of a single contained hydraulic fracture, like the ones
typically shown in Figs. 4.3, to develop an expression for predicting the stress shadow size,
as shown in Eq. (4.4):
_ , 10
1 , 10
(4.4)
The proposed Eq. (4.4) contains two coefficients as , and , which are
functions of Poisson’s ratio and stress anisotropy. It also contains multiplier , a
quadratic function of aspect ratio, which applies the effect of hydraulic fracture geometry
Based on these numerical results, it is implied that shadow and aperture of a single
hydraulic fracture is highly influenced by its containment.
4.5.4. Reliability of the proposed functions for uncontainment multiplier determination
In order to check the precision of the predicted uncontainment multipliers with respect to
the numerical ones, Figs. 4.13 are plotted which show the reliability of predictions using
these equations.
Figure 4. 13. Prediction of aperture and shadow size change by fracture uncontainment
1 The term “multiplier” in this thesis means that it only takes the effect of the extra variable (e.g. uncontainment, or simultaneous/multistage fracing) into account and it should be multiplied to the basic case (prediction without this variable) to have this extra effect incorporated in prediction of the parameter under study (aperture, shadow, or propagation potential).
y = 1.0006x + 0.0011R² = 0.9997
0.8
1
1.2
1.4
1.6
0.8 1 1.2 1.4 1.6
Predicted aperture ratio
Numerical aperture ratio
y = 0.923x + 0.0812R² = 0.9972
0.8
1
1.2
1.4
1.6
0.8 1 1.2 1.4 1.6
Predicted shad
ow ratio
Numerical shadow ratio (T=5°)
y = 1.0015x + 0.0009R² = 0.9992
0.8
1
1.2
1.4
1.6
0.8 1 1.2 1.4 1.6
Predicted shad
ow ratio
Numerical shadow ratio (T=80°)
57
As it can be seen from Figure 4.13, the ratio of aperture and shadow size of an
uncontained hydraulic fracture (penetrating into the bounding layers) over those of the
contained hydraulic fracture can be predicted using Eqs. (4.9-4.13).
4.5.5. Use of uncontainment multipliers (standalone fractures)
The aperture multiplier obtained from section 4.5.3 ( ) can be simply multiplied with
Eq. (4.1) for aperture prediction of a single uncontained hydraulic fracture. Similarly, for
shadow size change, the obtained multipliers of shadow in Section 4.5.3 (
and ) should be multiplied with Eqs. (4.4,4.8) respectively for prediction of Shadow
size for minimum and maximum threshold angles ( 5°, 80°).
4.6. Effect of simultaneous fracturing on shadow size and aperture
Simultaneous hydraulic fracturing of multilateral wells has been proven to be an efficient
way of having a better fracture network for hydrocarbon flow in unconventional shales. In
this part, two contained hydraulic fractures are assumed from two parallel horizontal wells
(see Figure 3.1; Scenario 3). It is assumed that there is no offset between the crack tips,
which means that the two fractures are aligned on a single plane and their distance is only
changed. The distance change has been assumed by changing of the distance between the
wells, not the aspect ratio or length of the fractures. Referring again to the presented
numerical scheme in Table 3.1, one can see that an extensive number of numerical analyses
are required for this purpose.
4.6.1. Change of stress shadow size by simultaneous fracturing ( 5°)
According to the fact that stress shadow has been satisfactorily predicted for a single
contained hydraulic fracture, the best approach was determined to merely compare the
58
shadow size of contained simultaneous fractures with that of a single contained fracture
(comparing Scenario 3 with Scenario 1; see Figure 3.1). Therefore, multiplying a multiplier
to the single fracture equation will result the shadow for simultaneous fractures.
In order to do so, the ratios of _ / _ and _ / _ were investigated
in detail in order to come up with the multiplying factors to be multiplied with the shadow
size around a contained fracture in a standalone well to predict the shadow size around
simultaneous contained fractures. It is worth indicating that _ defines the shadow size
for simultaneous fractures, while _ is the shadow size for a single fracture.
According to the numerical results, it was observed that hydraulic fracture aspect ratio
again plays an important role in the magnitude of multiplying factor for reduction of
shadow size around simultaneous fractures. From a statistical analysis on the ratio of
shadow size of simultaneous over single fractures, it was evident that this ratio is only
changing by the aspect ratio and distance between hydraulic fractures (for threshold angle
of ( 5°)). Figure 4.14 shows the results of this analysis for simultaneous fractures with
crack tip distances ranging from 0.07 to 3.0 .
Figure 4. 14. Shadow reduction by simultaneous fracturing
0
0.2
0.4
0.6
0.8
1
1.2
0 0.5 1 1.5 2 2.5 3
Mean Shad
ow ratio (SH
_D/SH_S)
Normalized tip distance, (x/c)
(AR=1.0)(AR=0.8)(AR=0.6)(AR=0.4)(AR=0.2)
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0 0.5 1 1.5 2 2.5 3
SD of Sh
adow ratio (SH
_D/SH_S)
Normalized tip distance, (x/c)
(AR=1.0)(AR=0.8)(AR=0.6)(AR=0.4)(AR=0.2)
59
As it can be seen from Figure 4.14, shadow ratio of _ / _ starts from 0.69 for a
normalized distance of 0.07 (aspect ratio of unity) and it becomes larger by decreasing the
aspect ratio of hydraulic fractures. As an example, shadow ratio starts from 0.83 for
normalized distance of 0.07 (aspect ratio of 0.2). Another important point about this
analysis is that approaching of shadow ratio to unity for hydraulic fractures with lower
aspect ratios is seen in lower distances between the tips. This means that the lower the
aspect ratio is, the closer the fracture tips must be together to have reduction effect of
stress shadow on each other.
Since these values are determined by a statistical analysis, it is beneficial to also report the
associated standard deviations (SD) for the analyses. As it is evident from Figure 14.4b,
SDs of the analyses are in between 0.015-0.060 and show a more or less decreasing trend
by increasing the distance between the hydraulic fractures.
4.6.2. Mathematical equation for prediction of simultaneous multiplier
After these interpretations, it is also required to have a robust equation for shadow
reduction ratio when we are dealing with simultaneous contained hydraulic fractures. It was
feasible to develop a function to predict the simultaneous multiplier with the numerical
results. Eq. (4.14) shows that the multiplier can be determined for all aspect ratios.
__
1 / (4.14)
It can be seen from Eq.(4.14) that there are two coefficients for this function, which can be
obtained for different aspect ratios from Table 4.5.
60
It is worth indicating that the coefficients for any aspect ratio can be determined via a
linear interpolation between those of the boundary aspect ratios.
Table 4. 5. Coefficient of the function in Eq. (4.14) AR 1.0 0.33522 1.057000.8 0.31354 1.402300.6 0.29420 1.997150.4 0.27786 3.412530.2 0.27772 7.38754
4.6.3. Reliability of the proposed equation for simultaneous multiplier prediction
In this section, the numerically obtained multipliers are compared with the mathematically
predicted ones using Eq. (4.14). The calculated simultaneous multipliers with the proposed
function are shown in Figure 4.15.
Figure 4. 15. Prediction of shadow size change by simultaneous hydraulic fracturing
It can be seen from Figure 4.15 that Eq. (4.14) predicts the shadow change satisfactorily. It
is important to notice that all these discussions are stress shadow change for threshold
angle of 5°. In the following section, change of minimum shadow size is investigated.
y = 1.0326x ‐ 0.0265R² = 0.9969
0.6
0.7
0.8
0.9
1
0.6 0.7 0.8 0.9 1
Predicted shad
ow chan
ge
Numerical Shadow change
61
4.6.4. Change of stress shadow size by simultaneous fracturing ( 80°)
According to the numerical results, it was observed that simultaneous hydraulic fracturing
of parallel laterals influences the minimum shadow size ( 80°) not as significantly as
that of the threshold angle of 5°. In fact, shadow size change for threshold angle of 80° is
much smaller compared to that for threshold angle of 5°. The results of this study are
shown in Table 4.6.
It is evident from Table 4.6 that the decrease in shadow is from 0.93 (for aspect ratio of
unity) to almost no change (for aspect ratio of 0.2).
It is worth mentioning that apertures of hydraulic fractures are not influenced by
simultaneous fracturing of parallel wells in case there is no overlap between the hydraulic
fractures. This means that maximum half-aperture of simultaneous fractures can be
determined via using the same equation for hydraulic fractures in standalone wells (see Eq.
between the tips. This means that as the offset between the tips increases, maximum SIF
change is seen at larger distances between the tips. The red line on Figure 5.10b shows the
maximum SIF change for different offset values. As it is seen, the highest SIF change is
moved to higher distances between the tips as the offset increases. The highlighted redline
in Figure 5.10b, is slightly shifted to smaller distances between the tips for lower aspect
ratios. Figure 5.11 shows this shifting of maximum values to lower distances between the
tips.
As it is evident from Figure 5.11, some data points are missing for lower aspect ratios. This
is because maximum values of SIF change for these missing points are located at distances
smaller than 0.1 .
Figure 5. 11. SIF maximum line for different aspect ratios
In addition, the gradient of the SIF change after its maximum value is higher for shorter
offset magnitudes. As the offset increases, gradient of SIF change decreases and this causes
the SIF change to be more for higher offset values at higher distances.
1
1.02
1.04
1.06
1.08
1.1
1.12
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
SIF ratio of simultan
eous/stan
dalone fractures
Distance between the tips, Ls/c
AR=1.0
AR=0.4
AR=0.6
AR=0.8
82
5.7. Use of the SIF change prediction functions
In the preceding sections, a complete set of equations were proposed for prediction of any
change in SIF as a result of using different fracturing technique/pattern. Using these sets
of equations, one can determine the SIF behavior corresponding to the selected fracing
strategy/pattern. Therefore, according to the availability of equipment and maximum
applicable pressure, the allowable SIF change is obtained and based on that, spacing
between adjacent hydraulic fractures is determined.
83
Chapter 6: Work flow for optimization of hydrofracing
6.1. General
In Chapters 4 and 5, the results of a comprehensive study for stress shadow, aperture, and
propagation potential of hydraulic fractures were presented. The results were qualitatively
and quantitatively discussed and complete sets of equations were proposed for prediction
of stress shadow, aperture, and the SIF change in hydraulic fractures. The reliability of all
the equations was also verified using the numerical results. In this chapter, the final work
flow for geomechanical optimization of hydraulic fracturing is proposed according to the
equations proposed in the two preceding chapters.
6.2. Successive procedure of hydraulic fracturing optimization
As indicated in Chapter 1, optimization of hydraulic fractures is performed from different
perspectives/approaches. Any of these perspectives are of paramount importance in their
own scale and influence on production and the cost for completion. In fact, they are not
separated from each other but instead they should be investigated together for an
optimized hydraulic fracturing. Four important steps for the optimization of hydraulic
fractures in unconventional shales are given below:
6.2.1. Step one
In the first step, optimization of hydraulic fracturing should be investigated from a
production point of view. In this method, porosity, permeability, fracture conductivity,
reservoir geometry and its fluid boundary conditions are used in a reservoir simulator for
optimization purpose. In this step, horizontal well spacing, fracturing pattern, and fracture
84
geometry are investigated and an optimized fracturing pattern including fracture geometries
are defined.
However, according to the fact that aperture of hydraulic fractures plays a key role in
conductivity of the fractures, it is first required to have an estimate about the fracture
aperture according to its influencing variables. Aperture is influenced by fracture geometry
(length and aspect ratio), moduli of the medium, excess pressure, and the used fracturing
technique (multistage fracing). From all the above-mentioned influencing factors in
aperture determination, except for the geometry of the fracture which should come from
the flow simulator, all the other variables are geomechanical factors defined in Chapter 4
and 5. The general equation for prediction of aperture for a hydraulic fracture is
determined as:
, , (6.1)
Estimation of aperture for a hydraulic fracture is done using the above equation together
with Eqs. (4.2), (4.3), (4.9), (4.12), (4.13), (5.1) and Tables 4.4, 5.1. Then, from the aperture,
proppant size/type, and finally, fracture conductivity is estimated. Therefore, in each try
for production optimization by changing fracture geometry/pattern, hydraulic conductivity
of the fractures should also be updated based on the new aperture.
6.2.2. Step two
In this step, having fracture geometry, stress shadow analysis can be used for determination
of shadow zone around the hydraulic fractures according to the in-situ stress regime, rock
moduli, and excess pressure by using the sets of equations proposed in Chapter 4. The
85
general form of shadow prediction for threshold angle of 5° is written as Eq. (6.2) which is
used for stress shadow size of simultaneous uncontained hydraulic fractures.
_ , 10
1 , 10
(6.2)
In this equation, Eqs. (4.5), (4.6), (4.7), (4.14), (4.15), (4.7), (4.12), (4.13) and Tables 4.1, 4.5,
4.7 are used to give the value for shadow size of a hydraulic fracture under simultaneous
uncontained hydraulic fracturing.
6.2.3. Step three
In this step, using the fracture geometry from step one, the allowable propagation potential
decrease (decrease of the SIF in the target zone; mentioned in Chapter 5), and considering
the fracing technique used for the job (simultaneous, multistage), the spacing between
hydraulic fractures are determined. For the case of simultaneous multistage fracturing of
rock, Eq. (5.3) together with Table 5.3 can be used for investigation of propagation
potential of this fracing technique.
6.2.4. Step four
In this final step, according to the fracture geometry from step one, and proposed
fracturing patterns from steps one, two, and three, a final optimized fracturing pattern is
obtained by making decision between the three fracturing patterns in the three preceding
steps.
86
Chapter 7: Concluding remarks and recommendations
Unconventional resources, having low permeability, are considered as potential
hydrocarbon reserves because of unconventional architecture; new drilling and completion
technology. Hydraulic fracturing plays a decisive role in economy of these reservoirs since
it makes the best productive regions connected to the wellbore, thereby raising the
permeability of the reservoir. In fracturing treatment, however, it is required to optimize
the fracing job in an efficient way to reduce the cost and make the best optimized fracture
network.
7.1. Aperture analysis
In this part of the thesis, a comprehensive equation was proposed for prediction of
aperture of hydraulic fractures with in-situ stress anisotropy, rock moduli, net pressure,
fracture aspect ratio, containment, and using different fracturing patterns (simultaneous
and/or multistage fracturing) as input variables. It was observed that stress anisotropy does
not have any effect on the aperture of hydraulic fractures. It was also observed that
aperture is inversely related to the rock moduli and directly related to the aspect ratio of the
fracture. Uncontainment of the hydraulic fracture highly influences the aperture of the
fracture, while simultaneous fracing does not have significant effect on the aperture size. In
case of lower Young’s modulus for the bounding layers and more penetration extent,
aperture increases. On the other hand, for the case of higher Young’s modulus for the
bounding layers and higher penetration, a lower aperture is observed. Finally, it was also
shown that aperture of the hydraulic fractures is negatively influenced by multistage
fracturing. In Chapter 5, this decrease in aperture was quantified considering different
spacing between the fractures.
87
7.2. Stress shadow analysis
First, it was shown that shadow mechanism of hydraulic fractures is different depending on
the net pressure magnitude. In addition, shadow size around a hydraulic fracture was also
calculated assuming the same variables used in aperture determination. It was observed that
Young’s modulus of the rock (using constant pressure boundary condition) does not
change the shadow size. Stress anisotropy, Poisson’s ratio, and net pressure directly
increase the shadow size. Effect of simultaneous fracturing and uncontainment were also
calculated as multipliers to be multiplied with the case of single/contained fractures for
shadow determination for these scenarios. A comprehensive set of equations for shadow
size was also proposed by which shadow size of hydraulic fractures can be predicted
satisfactorily. These equations are really useful in hydraulic fracturing treatment and design
in term of perforation distance and proppant size/type for the fracture.
7.3. Analysis of propagation potential
Based on the consideration that stress intensity factor (SIF) defines the propagation
potential of a hydraulic fracture, a comprehensive numerical simulation framework was
designed in chapter 5 to investigate the interaction between fracture tips and the influence
on the SIF. Different scenarios were considered specifically, fracture aspect ratio, spacing
between multistage fractures, and distance and offset between the tips in simultaneous
fracturing mode. First, it was shown that multistage fracturing in a standalone well
dominantly reduces the SIF of the propagating fractures. The level of this decrease is lower
for the lower aspect ratios. The effect of spacing on the SIF was quantified using a fitting
equation with its proposed coefficients. Secondly, effect of two hydraulic fractures from
two parallel wells (single stage fracturing) was considered to study the effect of
88
simultaneous fracturing of parallel wells on the SIF change. It was observed that SIF of the
meeting hydraulic fractures increases noticeably as a result of simultaneous fracturing. The
magnitude of this change, however, is higher for higher aspect ratios. This effect was also
quantified proposing an equation and its coefficients. Thirdly, the effect of having
simultaneous multistage fracturing of parallel wells were compared with the case of single
stage fracturing of standalone wells in order to quantify the effect of simultaneous
fracturing when multistage fracturing technique is applied. Finally, in order to show the
effect of offset between the fracture tips, effect of offset/distance between the tips were
studied and the behavior of SIF as a result such pattern was also studied. It was observed
that existence of offset between fracture tips is a retarding factor in the SIF increase in
simultaneous fracturing. In addition, for simultaneous fractures with no offset, the lower
the distance between the tips, the higher the SIF increase. For the case of existing offset,
on the other hand, in case offset is more than one-fourth of the fracture length, the highest
SIF change no longer belongs to the least distance and there will be a certain distance in
which SIF is maximum. It is worth mentioning that the range of SIF change in this range
of offset (offset 0.4 ), is between 1.1 to 1.0 of a single stage fracture in a standalone well.
7.4. Optimized fracture network
An optimized fracture pattern was defined in Chapter 1 as “parallel fractures having the
highest production, the highest propagation potential in the target formation, perpendicular
to the wellbore axis, with an optimized distance to prevent any deviation/collapse”.
89
7.5. Four steps in optimization of hydraulic fracturing
In order to reach the goal of optimized fracturing pattern, three important aspects from
geomechanical and production perspectives should be considered simultaneously to come
up with the final decision on the fracturing pattern.
It has been stated in the literature that low permeability of the reservoir makes the
fracturing of it essential to produce at an economical rate. The lower the permeability, the
higher the number of fractures should be generated to have efficient production (Soliman
et al., 1997). On the other hand, fractures cannot be placed too close to each other because
of the geomechanical aspect of fracturing since propagation potential may substantially
decrease and fractures may not propagate in the direction perpendicular to the wellbore.
Therefore, the following four steps are suggested in this thesis for an efficient optimization
of hydraulic fractures considering the most influencing parameters into account.
1. In the first step, a primary fracturing pattern is assumed based on experience or
data from adjacent wells. Then, calculating the aperture of the fractures from the
equations proposed in Chapter 4, proppant type/size is selected and estimation on
the fracture conductivity is made. After that, fracturing pattern and fracture
geometry is optimized noting the fact that in each pattern new fracture conductivity
should be calculated based on the new aperture. Finally, in this step, an initial
estimate is given on fracture geometry and fracturing pattern.
2. In the second step, according to the geomechanical data and the obtained fracture
geometry from the first step, stress shadow size around hydraulic fractures with
their own fracing pattern is estimated using the set of equations proposed in
90
Chapter 4. In this step, the second estimate about fracturing pattern; distance
between parallel wells and spacing between multistage fractures is made.
3. In the third step, the estimated fracturing pattern in the second step is used for
determination of the change in propagation potential of the hydraulic fractures
compared to the case of a single staged fracture in a standalone well.
4. In this final step, according to the obtained propagation potential change in the
third step and considering the allowable propagation potential decrease, the best
spacing between adjacent fractures in multistage fracturing and the best distance
between parallel wells is estimated based on the two estimations made in the first
and second steps. Basically, maximum distance obtained based on economical
production (first step), no deviation or collapse (second step), and allowable
propagation potential decrease (third step) is selected as the optimum spacing
between the fractures.
In this thesis, a detailed qualitative and quantitative study of aperture of a hydraulic fracture
required for the first step, stress shadow size for the second step, and change in
propagation potential in the third steps were done in order to come up with a complete
framework for optimization of fracture treatment in unconventional shales.
91
References
1. Abousleiman Y., Tran M., Hoang S., Bobko C., Ortega A., Ulm G.J. 2007. Geomechanics field and laboratory characterization of Woodford shale: The next gas play. Paper SPE 110120, presented at SPE annual technical conference and exhibition, Anheim, CA, USA. 11-14 November.
2. Andrews A., Folger P., Humphries M., Copeland C., Tiemann M., Meltz R., and Brougher C. 2009. Unconventional Gas Shales: Development, Technology, and Policy Issues. Congressional research service.
3. Atkinson, C., Smelser, R.E. & Sanchez, J. 1982. Combined mode fracture via the cracked Brazilian disk test. Int. J. Fract.; 18: 279-291.
4. Barker, L.M. 1977. A simplified method for measuring plane strain fracture toughness. Eng. Fract. Mech.; 9: 361-369.
5. Barree R.D., Gilbert JV., Conway M.W. 2009. “Stress and rock property profiling for unconventional reservoir stimulation. SPE 118703, presented at SPE hydraulic fracturing technology conference, The Woodlands, TX, USA, 19-21 January.
6. Britt L.K., Schoeffler J. 2009. “The geomechanics of a shale play: what makes a shale prospective!” Paper SPE 125525, presented at SPE Eastern regional meeting, Charleston, WVA, USA, 23-25 September.
7. Broek D. 1982. Elementary Engineering Fracture Mechanics. Third Revised Edition. Martinus Nijhoff Publishers, Boston, USA.
8. Chang, S.-H., Lee, C.-I. & Jeon, S. 2002. Measurement of rock fracture toughness under modes I and II and mixed-mode conditions by using disc-type specimens. Eng. Geol.; 66: 79-97.
9. Cheng Y. 2009. Boundary Element Analysis of the Stress Distribution around Multiple Fractures: Implications for the Spacing of Perforation Clusters of Hydraulically Fractured Horizontal Wells. SPE 125769.
10. Chong, K.P. & Kuruppu, M.D. 1984. New specimen for fracture toughness determination of rock and other materials. Int. J. Fract.; 26: 59-62.
11. Daneshy A., Au-Yeung J., Thompson T., Tymko D. 2012. Fracture shadowing: A direct method for determination of the reach and propagation pattern of hydraulic fractures in horizontal wells. SPE 151980.
12. Dusseault M., McLennan J. 2011. Massive Multistage Hydraulic Fracturing: Where are We? ARMA (American Rock Mechanics Association) e-Newsletter. Published, 01/2011.
92
13. Evans, A.G. 1972. A method for evaluating the time dependent failure characteristics
of brittle materials and its application to polycrystalline alumina. J. Mater. Sci.; 7: 1137-1146.
14. Fischer M.P., Gross M.R., Engelder T., Greenfield R.J. 1995. Finite-element analysis of the stress distribution around a pressurized crack in a layered elastic medium: implications for the spacing of fluid-driven joints in bedded sedimentary rock. Tectonophysics 247, 49-64.
15. Fisher, M.K., Heinze, J.R., Harris, C.D., Davidson, B.M, Wright, C.A., and Dunn, K.P. 2004. “Optimizing Horizontal Completion Techniques in the Barnett Shale Using Microseismic Fracture Mapping”, paper SPE 90051 presented at the SPE Annual Technical Conference and Exhibition, Houston, Texas, September 26-29.
16. Fowell, R.J. 1995. Suggested methods for determining Mode I fracture toughness using cracked chevron notched Brazilian disc specimens. Int. J. Rock Mech. Min. Sci. & Geomech. Abstr.; 32: 57-64.
17. Green, A.E., Sneddon, I.N. 1950. “The distribution of stress in the neighborhood of a flat elliptical crack in an elastic solid”. Mathematical Proceedings of the Cambridge Philosophical Society. 46(01): p. 159-163.
18. Guidera J.T., and Lardner R.W. 1975. Journal of Elasticity, 5, 59-73.
19. Guo, H., Aziz, N.I. & Schmidt, L.C. 1993. Rock fracture toughness determination by the Brazilian test. Eng. Geol.; 33: 177-188.
20. Hay J, Sondergeld C.H. 2010. Mechanical Testing of Shale by Instrumented Indentation. Application Note. Agilent Technologies, Inc. Printed in USA, May 5. 5990-5816EN.
21. Inglis, C.E. 1913. Stresses in a plate due to the presence of cracks and sharp corners. Transactions of the Institution of Naval Architects, Vol. 55, pp. 219-230.
22. Irwin G. 1957. Analysis of stresses and strains near the end of a crack traversing a plate, Journal of Applied Mechanics 24, 361–364.
23. Janssen M., Zuidema J., Wanhill R. 2009. Fracture Mechanics: Fundamentals and Applications. Taylor and Francis.
24. Jeffrey, R.G. 1989. The Combined Effect of Fluid Lag and Fracture Toughness on Hydraulic Fracture Propagation. SPE 18957.
25. Jo H. 2012. Optimizing Fracture Spacing to Induce Complex Fractures in a Hydraulically Fractured Horizontal Wellbore. SPE 154930.
93
26. Kassir, M. K. 1981. Stress-intensity factor for a three-dimensional rectangular crack. ASME J. Appl. Mech. 48, 309-312.
27. Keer, L.M. 1964. Quarterly Journal of Mechanics and Applied Mathematics, 17, 423-436.
28. King G.E. 2010. Thirty years of gas shale fracturing: what have we learned?. Presented at the SPE annual technical conference and exhibition held in Florence, Italy, 19-22 September.
29. King G.E., Haile L., Shuss J., and Dobkins T.A. 2008. Increasing Fracture Path Complexity and Controlling Downward Fracture Growth in the Barnett Shale. SPE 119896.
30. Kirsch. 1898. Die Theorie der Elastizität und die Bedürfnisse der Festigkeitslehre. Zeitschrift des Vereines deutscher Ingenieure, 42, 797–807.
31. Mastrojannis E.N., Keer L.M., Mura, T. 1979. Stress intensity factor for a plane crack under normal pressure. International Journal of Fracture, Vol. 15, No. 3, June.
32. Morrill, J.C. and Miskimins, J.L. 2012. Optimizing Hydraulic Fracture Spacing in Unconventional Shales, SPE 152595, SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, February 5-8.
34. Muskhelishvili, N.I. 1953. Some basic problems of mathematical theory of elasticity (1933). English translation, Noordhoff.
35. Mutalik P.N., Gibson B. 2008. Case history of sequential and simultaneous fracturing of the Barnett shale in Parker county. Presented at the 2008 SPE annual technical conference and exhibition held in Denver, USA, 21-24 September.
36. Nicolas P. Roussel, SPE, and Mukul M. Sharma. 2011a. Strategies to Minimize Frac Spacing and Stimulate Natural Fractures in Horizontal Completions. SPE 146104.
37. Ouchterlony, F. 1988. Suggested methods for determining the fracture toughness of rock. Int. J. Rock Mech. Min. Sci. & Geomech. Abstr.; 25: 71-96.
38. Paris P.C. and Sih G.C. 1965. Stress analysis of cracks. ASTM STP 381, pp. 30-81.
39. Eshelby, J.D. 1968. Stress analysis: Elasticity and fracture mechanics, ISI publications, 121, pp. 13-48.
40. Roussel N.P., and Sharma M. 2011b. Strategies to Minimize Frac Spacing and Stimulate Natural Fractures in Horizontal Completions. SPE 146104.
94
41. Rafiee M., Soliman M.Y., Pirayesh E. 2012. Hydraulic fracturing design and optimization: A modification to zipper frac. SPE 159786.
42. Shah R.C. and Kobayashi A.S. 1971. Engineering Fracture Mechanics, 3, 71-96.
43. Sih, G.C. 1973. Handbook of Stress-intensity Factors: Stress-intensity Factor : Solutions and Formulas for References. Institute of Fracture and Solid Mechanics, Lehigh University.
44. Singh I, Miskimins J.L. 2010. A Numerical Study of the Effects of Packer-Induced Stresses and Stress Shadowing on Fracture Initiation and Stimulation of Horizontal Wells. CSUG/SPE 136856. Presentation at the Canadian Unconventional Resources & International Petroleum Conference held in Calgary, Alberta, Canada, 19–21 October.
45. Sneddon, I.N. and Elliot, H.A. 1946. The Opening of a Griffith Crack under Internal Pressure. Quarterly of Applied Mathematics 4 (3): 262–267.
46. Sneddon I.N. and Lowengrub M. 1969. Crack Problems in the Classical Theory of Elasticity, J. Wiley and Sons, New York, N.Y.
47. Soliman M. Y., Azari M., Hunt J. L., Chen C.C. 1996. Design and Analysis Of Fractured Horizontal Wells in gas reservoirs. SPE 35343.
48. Soliman M.Y., Boonen P. 1997. Review of fractured horizontal wells technology. SPE 36289.
49. Sun, Z. & Ouchterlony, F. 1986. Fracture toughness of stripa granite cores. Int. J. Rock Mech. Min. Sci. & Geomech. Abstr.; 23: 399-409.
50. Tada H., Paris P.C., Irwin G.R. 2000. The stress analysis of cracks handbook. Third edition. ASME press, New York.
51. US Energy Information Administration. 2011. Annual energy outlook. (http://www.eia.gov).
52. Warpinski, N.R., Branagan, P.T., Peterson, R.E., Fix, J.E., Uhl, J.E., Engler, B.P., and Wilmer, R. 1997. “Microseismic and Deformation Imaging of Hydraulic Fracture Growth and Geometry in the C Sand Interval, GRI/DOE M-Site Project”, paper SPE 38573 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, October 5-8.
53. Waters G., Dean B., and Downie R., Kerrihard K., Austbo L., and McPherson B. 2009. Simultaneous Hydraulic Fracturing of Adjacent Horizontal Wells in the Woodford Shale. SPE 119635.
54. Westergaard H.M. 1939. Bearing pressures and cracks. Journal of applied mechanics. 61, pp. A49-A53.
95
55. Whittaker, B.N., Singh, R.N. & Sun, G. 1992. Rock Fracture Mechanics, Principles, Design and Applications. Developments in Geotechnical Engineering, 71. Elsevier, Amsterdam.
56. Willis, L.R. 1968. International Journal of Engineering Science, 6, 253-263.
57. Wong S.W., Geilikman M., Xu G. 2013. Interaction of multiple hydraulic fractures in horizontal wells. SPE 163982.
58. Xu W., Calvez J.L., Thiercelin M. 2009. “Characterization of hydraulically-induced fracture network using treatment and microseismic data in a tight gas formation: A geomechanical approach” Abstract SPE 125237 and slides, presented at SPE tight gas completions conference San Antonio, TX, USA. 15-17 June.