(/065/5 352&((',1*6 

UDC 52.47.19

GEOMECHANICAL STABILITY ANALYSIS FOR SELECTING

WELLBORE TRAJECTORY AND PREDICTING
SAND PRODUCTION

Phan Ngoc Trung, Nguyen The Duc, Nguyen Minh Quy

(Vietnam Petroleum Institute)

Geomechanical stability plays an important role in the development of long and deep wells. Borehole
collapse, circulation losses and sand production are costly problems for the petroleum production.
In the study presented here, a model based on Mohr-Coulomb failure criterion is used to analyze
wellbore stability for three synthetic cases with different stress regimes. For each case, the analyses are
performed to select wellbore inclination and azimuth for instability minimization. After the most stable
well direction is selected, the analyses are carried out to determine free-sanding bottomhole flowing
pressure (BHP) associated with different values of reservoir pressure in order to predict potential of
sanding in the future production process. The study shows that geomechanical stability analysis can
provide valuable supports for selecting wellbore trajectory and controlling sand production.
Keywords: water flooding, optimization algorithm, well rate allocation, artificial neural network,
RESERVOIR AND PETROLEUM ENGINEERING

water cut.
Adress: trungpn@vpi.pvn.vn
DOI: 10.5510/OGP20100400040

1. INTRODUCTION stability. There also are various failure criteria which

In the last two decades, the petroleum industry has
witnessed what can be called geomechanics revolution
and petroleum geomechanics has become the fastest
growing commercial area for technical investment
within the service sector [1]. Geomechanical stability
has become regular consideration from oil exploration
to production. The geomechanical instability is usually
faced in the drilling with high rig rates in deep
water, the drilling in tectonic fields, salt-domes, high-
pressure high-temperature fields, and the drilling of
more horizontal, highly deviated and multilateral wells
([2]-[4]). Another problem requiring geomechanical
stability analysis is related to sand production ([5]-
[7]). Production of reservoir fluids at high rates (low
bottomhole flowing pressure) cause an increase in the
induced tangential stresses concentrated on the face
of an open hole or on the walls of perforations in a
cased hole. If these induced stresses exceed formation
in situ strength, the formation will fail and sand
could be produced together with fluids of reservoir.
Therefore, sanding prediction needs a knowledge about
the mechanisms upon which the rock failure has
occurred. It is very important to exactly determine
what mechanism has caused the problem of formation
instability.
Instability of formation around a borehole
(or perforation tunnel) is usually evaluated with a
combination of constitutive models and failure criteria
([2], [8], [9]). Constitutive models are a set of equations
used to determine the stresses around the hole. They
range from simple linear elastic models to sophisticated
poro-elasto-plastic models. All the constitutive models
have only studied the effect of a few parameters
on the hole stability and have ignored the rest ([8]-
[11]). Actually, there is no constitutive model which
can handle all the parameters that affect the hole
are used to determine the onset of failure in the rocks.
Among them, the Mohr-Coulomb criterion is the most
common failure criterion encountered in geotechnical
engineering. Many geotechnical analysis methods and
programs require use of this failure criterion.
In this study, stability analyses have been performed
by using a combination of linear elastic constitutive
model and Mohr- Coulomb failure criteria. The method
has been employed to analyze wellbore stability for
three synthetic cases with different stress regimes. The
calculated results show the effect of inclination and
azimuth on wellbore stability is strongly dependent on
in-situ stress state. For the most stable wellbore of each
case, the analyses are also carried out for examining
the influence of reservoir depletion on the potential of
sanding. The study has demonstrated the important
role of geomechanical stability analysis in solution of
some practical problems in petroleum engineering.
2. DESCRITION OF ANALYTICAL METHOD
2.1. Stresses around hole
The holes of wellbore (or perforation tunnel) and
their adjacent formation are often approximated as
thick-walled hollow cylinder. Therefore, it is possible
to obtain a solution for the near hole stress state and
use it in stability analysis.
Assume that the principal stresses in the virgin
formation are: , the vertical stress, H the largest
horizontal stress, and h, the smallest horizontal stress.
A coordinate system (x', y', z') is oriented so that x' is
parallel to H , y' is parallel to h , and z' is parallel to 
(i.e. z'-axis is vertical; fig.1). The stresses in the vicinity
of the deviated hole are most conveniently described in a
coordinate system (x, y, z,) where the z-axis is parallel to
the hole, y-axis to be horizontal, and x-axis to be parallel
to the lowermost radial direction of the hole (fig.1).

24
 (/065/5 352&((',1*6 

lxx' = cos cos lxy' = -sin lxz' = sin cos

lyx' = cos sin lyy' = cos lyz' = sin sin (1)
lzx' = sin lzy' = 0 lzz' = cos

By transforming to the (x, y, z) coordinate system, the

formation stresses H, h and  become:

V x0 2
lxx 'V H  lxy 'V h  lxz 'V v
2 2

V y0 2
lyx 'V H  lyy 'V h  lyz 'V v
2 2
(2)
V z0 2
lzx 'V H  lzy 'V h  lzz 'V v
2 2

Fig.1. Coordinate system for a hole [2] W xy

As can be seen in Figure 2, a coordinate transformation
from system (x', y', z') to system (x, y, z) can be obtained
by two operations: 1) a rotation around z'-axis, and 2) a
rotation around the y-axis. The angle represents the hole
inclination and the angle represents the azimuth angle.
The transformation can be described mathematically
0
lxx 'lyx 'V H  lxy 'lyy 'V h  lxz 'lyz 'V v
W 0
yz lyx 'lzx 'V H  lyy 'lzy 'V h  lyz 'lzz 'V v
W 0
zx lzx 'lxx 'V H  lzy 'lxy 'V h  lzz 'lxz 'V v
Here the superscript 0 indicate that these are the
virgin formation stresses. Equations (2) represent the
stress state in the case of no hole in the formation.

RESERVOIR AND PETROLEUM ENGINEERING

by the following direction cosines:
lxx', lxy', lxz' - The cosines of the angles between x-axis and
x', y', z'-axes, respectively.
lyx', lyy', lyz' - The cosines of the angles between y-axis and
x', y', z'-axes, respectively.
lzx', lzy', lzz' - The cosines of the angles between z-axis and
x', y', z'-axes, respectively.
These cosines are related to the inclination angle and
the azimuth angle as:
The stress state will change when a hole exists in
the formation. For the case of cylindrical hole, it
is convenient to present the stresses in cylindrical
coordinate (r, , z). By assuming that there is no
displacement along z-axis (plane strain condition), a
derivation of the stress solution around cylindrical hole
can be found and the stresses at the hole wall are given
by the following equations:
Vr pw
VT  
V x0  V y0  2 V x0  V y0 cos  2T   4W xy
0
sin  2T   pw

 
V z V z0 Q 2 V x0  V y0 cos  2T   4W xy
0
sin  2T  
W rT 0 (3)
WT z 2W sin T  2W cos T
0
xz
0
yz

W rz 0
where pW is pressure at the wall of hole,  is Poisons
ratio and  indicate the angular position around the
hole (fig.2).
As failure is governed by the principal stresses i, j, k,
the following matrix equation defines planes of principal
stress:
V r 0 0 V i 0 0

0 VT WT z 0 V j 0
(4)
0 W T z V z 0 0 V
k

Taking the determinant of the above matrices, the

principal stresses are given by the following eigenvalue
equation:
^
V r  V  V T  V V z  V   WT2z (5) `
By solving above equation, the principal stresses acting
on the hole wall are given as:
Vi pw
1
V  V z   21 V T  V z 
2 (6)
Vj  4W T2z
2 T
1
V  V z   21 V T  V z 
2
Vk  4W T2z
2 T

Fig.2. Coordinate transformation [2] and the maximum and minimum stresses acting on the

25
 (/065/5 352&((',1*6 

hole wall will be as follow:

V1 max V i , V j , V k
(7)
V3 max V i ,V j ,V k

2.2. Failure criterion

For evaluating collapse of hole wall, the Mohr-
Coulomb failure criterion is employed (for example,
see [2], [3], [6]). This is governed by the maximum
and the minimum stresses. Fig.3 shows the Mohr-
Coulomb criterion and a Mohrs circle that touch the
failure line.
The Mohr-Coulomb criterion can be expressed Fig.3. Mohr-Coulomb failure criterion in  -  space
mathematically as follows:
most authors. The analysis of available FIT/LOT data
 = 0 + tan (8) suggested that the minimum horizontal stress equal to
9036 psi. However, no information can be employed
where,  and  are shear and normal stresses to exactly determine the maximum horizontal stress.
respectively, 0 is the inherent cohesion and  is the angle In order to cover potential uncertainty range, analyses
RESERVOIR AND PETROLEUM ENGINEERING

of internal friction. have been performed for three synthetic cases with
The shear and normal stresses can be calculated as, different maximum horizontal stresses:
H = 1.1 h = 9940 psi
 
1 ' 1. Base case:
W V 1  V 3' cos I (9)
2
2. Low stress case: H = h = 9936 psi
V 
1 ' 1
  
V  V 3'  V 1'  V 3' sin I
2 1 2
3. High stress case: H = 1.2 h = 13147 psi
where, '1 and '3 are maximum and minimum effective
stresses which can be calculated as: It should be noted that the stress state is usually
V 1' V 1  D p0 classified into three different stress regimes based on the
(10) relative magnitude between the vertical and horizontal
V 3' V 3  D p0
stresses (see [2], [12]). Normal or extensional faulting
where, p0 is pore pressure and  is Biots coefficient. (NF) stress regimes are associated with  H h ,
Combining the equations above, the failure condition reverse or compressional faulting (RF) stress regimes
becomes: are associated with H h  , and strike-slip (SS) stress
V 1'  V 3'  V 1'  V 3' sin I 2W 0 cos I (11)
According to Equation (6), in the case of collapse of
wellbore or perforation tunnel at low hole pressures, j
will be the maximum principal stress 1, and i will be the
minimum principal stress 3.
regimes are associated with H  h. According to the
classification, the base case and the low stress case are in
NF stress regime and the high stress case is in RF stress
5700

2.3. Computer program 5500

Azi. = 0 deg.
The modeling method described above have been 5300 Azi. = 30 deg.
used to write a computer program (using FORTRAN
Minimum BHP, psi

programming language) which is able to predicted 5100 Azi. = 60 deg.

collapse condition of the hole wall for any combination 4900 Azi. = 90 deg.
of in-situ stress state and pore pressure. The calculation
requires values of the following input parameters 4700
at the depth of the studied formation: (a) the in situ
4500
stresses and pore pressure, (b) the cohesion, internal
friction angle and Poissons ratio, and (c) the wellbore 4300
inclination and azimuth.
4100

3. CALCULATED RESULTS 3900

3.1. Description of synthetic cases
3700
Measured data from a field of Vietnam are used in
the synthetic cases: The sandstone has a cohesion of 0 10 20 30 40 50 60 70 80 90
1783 psi, a friction angle of 44.2 degree, and a Poisons
ratio of 0.15. At a production depth of 11142 ft, the Inclination, degree
vertical stress is equivalent to the overburden pressure,
equal to 10956 psi, the pore pressure is taken at 4836 Fig.4. Critical Bottomhole Pressure as functions
psi, and the Biots factor is set to 0.7 as suggested by of inclination (base case)

26
 (/065/5 352&((',1*6 

wellbore is a 40o-deviated one and in a plane parallel to

5700 the minimum in situ stress h.
The calculations of minimum bottomhole pressure
5500 for the low stress case are presented in Figure 5 for
5300 different wellbore inclination and azimuths. Because
Minimum BHP, psi

of the isotropic horizontal stress state of this case, the

5100
results should be independent of wellbore azimuth
4900 angle. This expectation is clearly shown in Figure 5
4700 where plots associated with different azimuths are in
the same. For this case, the most stable trajectory is
4500 exactly vertical, that is inclination angle = 0o.
4300 Figure 6 presents calculated results for the high stress
4100
case. The case is in an RF stress regime with anisotropic
horizontal stress. Contrary to two above cases, the
3900 most stable wellbore inclination is horizontal. The most
3700 stable wellbore trajectory is associated with a horizontal
wellbore which has the azimuth angle equal to 30o.
0 10 20 30 40 50 60 70 80 90
In summary, the study on the effect of wellbore
Inclination, degree
inclination and azimuth indicates that: vertical boreholes
will minimize the potential borehole instability only
when the stress state is horizontally isotropic and in NF

RESERVOIR AND PETROLEUM ENGINEERING

Fig.5. Critical Bottomhole Pressure as functions of stress regime. Having anisotropic horizontal stress and/
inclination (low stress case) or being in RF stress regime will divert the most stable

regime. The difference between the base case and the

low stress case is that the first is in isotropic horizontal operating envelope
stress state while the second is in the stress state of of sand free
horizontal anisotropy. 5000
4500
3.2. Effect of wellbore inclination and azimuth 4000
The program has been used to study influence 3500
of inclination and azimuth on wellbore stability.
BHP, psi

3000
The minimum bottomhole flowing pressures (BHP)
2500
for wellbore stability are calculated with different
2000 sand failure
inclinations () and azimuths (). The results are shown
1500
in Figures 4-6.
1000
From the calculated results of the base case presented
500
in Figure 4, it is apparent that a vertical wellbore
0
is more stable than a horizontal wellbore with all
azimuths. However, the optimum drilling trajectory 0 1000 2000 3000 4000 5000
is not necessarily vertical. In this case, the most stable
Reservoir Pressure, psi

5700
Fig.7. Sand free operating envelope plot (base case)
5500
5300
Minimum BHP, psi

well path from the vertical direction. In these situations,

5100
deviated and horizontal wellbores are potentially
4900 more stable than vertical wellbores. The inclination
4700 and azimuth of the most stable wellbore should be
Azi.=0 deg. determined exactly by geomechanical stability analyses.
4500 3.3. Effect of reservoir pressure depletion
Azi=30 deg.
4300 The aforementioned calculations are obtained with
Azi.=60 deg. the initial reservoir (pore) pressure. However, the
4100
Azi.=90 deg. reservoir pressure may be decreased during production
3900 process. In order to show the influence of reservoir
3700 depletion, the analyses have been carried out for these
three cases with different reservoir pressures. For each
0 10 20 30 40 50 60 70 80 90
case, the most stable wellbore trajectory (inclination and
azimuth) is used in the calculation. The obtained results
Inclination, degree
for base case, low stress case, and high stress case are
shown in Figures 7-9, respectively. For these figures,
Fig.6. Critical Bottomhole Pressure as functions of it should be noted that the bottomhole pressure must
inclination (high stress case) be lower than reservoir pressure in a production well.

27
 (/065/5 352&((',1*6 

case is seen in Figure 7. As the reservoir pressure

operating decreases from 4836 psi (initial reservoir pressure) to
envelope of sand 3800 psi, the minimum bottomhole pressure of sand
5000 free production decreases from 4108 psi to 3800 psi (i.e.
4500 maximum drawdown pressure decreases from 728 psi
4000 to 0 psi). It means that the well can not produce without
3500
sand failure when the reservoir pressure decreases
below 3800 psi.
BHP, psi

3000
Figure 8 shows the sand free operating envelope
2500
plot for the low stress case. As the reservoir pressure
2000 sand failure decreases from 4836 psi to 2800 psi, the minimum
1500
bottomhole pressure decreases from 3818 psi to 2800
1000
psi (i.e. maximum drawdown pressure decreases from
500
1018 psi to 0 psi). It means that the well can not produce
0 without sand failure when the reservoir pressure below
0 1000 2000 3000 4000 5000 2800 psi. The sand free production period in this case is
therefore can be longer than in the base case.
Reservoir Pressure, psi For the high stress case, the sand free operating
envelope plot is presented in Figure 9. At the initial
Fig.8. Sand free operating envelope plot reservoir pressure of 4836 psi, the minimum bottomhole
(low stress case) pressure is equal to 4534 psi. The well can not produce
RESERVOIR AND PETROLEUM ENGINEERING

without sand failure when the reservoir pressure below

Therefore the operating points must be in the lower- 4200 psi. It means that the operating envelop of sand
right half part of the graph. This part is then divided free production in this case is much smaller than the
into sand free operating envelope and sand failure zone. ones in two previous cases.
The sand free operating envelope plot for the base
4. CONCLUSION
operating envelope of A method for analyzing geomechanical stability of
sand free produc tion the holes (open hole or perforation tunnel in cased hole)
has been presented.
5000 Wellbore stability analyses using the presented
4500 method have been performed for some synthetic
4000 cases. The obtained results show the influences of well
3500 inclination, well azimuth, and reservoir depletion under
different stress regime.
BHP, psi

3000
 The presented study results shows methodology
2500
can be employed in:
2000 sand failure zone  Predicting onset of sanding production for
1500 existing free-sanding well.
1000  Determining optimum drawdown for existing
500 sanding well.
0  Optimizing wellbore trajectory/perforation
direction to minimize instability problem for future
0 1000 2000 3000 4000 5000 infill well.
Reservoir Pressure, psi In order to improve the accuracy of the predictions,
more works should be carried out for modeling the effect
Fig.9. Sand free operating envelope plot of water-cut increase, the effect of high compressibility
(high stress case) of production fluid in gas producer, etc.

References

1. P.Papanastasious, A.Zevos. Application of Computational Geomechanics in Petroleum Engineering

//Proceeding of 5th GRACM International Congress on Computational Mechanics. Limassol. 2005.
2. E.Fjaer. Petroleum Related Rock Mechanics. Elsevier Publications, 1992.
3. S.Zhou, R.R.Hillis, M.Sandiford. On the Mechanical Stability of Inclined Wellbore //SPE Drilling &
Completion. -1996. No.2. P.67-73.
4. E.Karstad, B.S.Aadnoy. Optimization of Borehole Stability Using 3D Stress Optimization //SPE Annual
Technical Conference and Exhibition. Texas: Dallas. -2005. Paper 97149-MS.
5. W.L.Penperthy, C.M.Shaughnessy. Sand Control //SPE. Texas: Richarson. 1992.
6. X.Yi, P.P.Valko, J.Russel. Effect of Rock Strength Criterion on the Predicted Onset of Sand Production
//International Journal of Geomechanics. -2005. No.3 -P.66-73.

28
(/065/5 352&((',1*6 

7. M.N.J.Al-Awad, S.F.M.Desouky. Prediction of Sand Production from a Saudi Sandstone Reservoir

//Oil Gas Science and Technology. -1997. -V.52. -No.4. -P. 1-8.
8. J.A.Hudson, J.P.Harrison. Engineering Rock Mechanics: An Introduction to the Principles. New
York: Pergamon Press Inc., 1997
9. P.A.Chales, S.Roatesi. A Fully Analytical Solution of Wellbore Stability Problem under Undrained
Condition Using a Linearised Cam-Clay Model //Oil Gas Science and Technology. -1999. -V.54. -No.5.
-P.551-563.
10. B.Vasarhelyi, P.Van. Influence of Water Content on the Streng of Rock //Engineering Geology.
-2006. -V.84. -P.70-74.
11. B.Wu, C.P.Tan. Sand Production Prediction of Gas Field: Methodology and Laboratory
Verification //SPE ISRM Rock Mechanics Conference. Irving, Texas: Irving. -2002. Paper 77234.
12. E.M.Anderson. The Dynamics of Faulting and Dyke Formation with Applications to Britain.
Edinburgh: Oliver and Boyd, 1951







RESERVOIR AND PETROLEUM ENGINEERING

 



   


! 
" #

  



$
%




 
  
 !
" #  $ 

    
 
 % !
" *
< ! %>?
"  K

* 
!
  !
" %K $ Q Q  $ $!  $!<  * 



$ <% >  <!
K $<
!  ! <
 
 <K  !



  
*  X
YZ
 \^_`Y{^|}^~K $  <



! !
  !
" % <   ! % 
!  
 % "


$" >
"<


$!<<!%





<
 *  !>  !%

 
 
$
!   !
" %
$!<

  Q# $<    
  <
!  !% 
 $
K
!

 %
 $
!<
! <$ !
$ Q


$ $!  ! < $Q <%> <!
 $
%!
K 



  
  " 

 Q # $ $ !% 
 
!
 !
" %  !% 
$
.

&'*'+/+0;<=<=>?@*VX>Z?<*@;\=\=;V^<+_0;<`0j'_>0k0q/?/=/=
x?Zj=Zk'/^/=zVZ_V{@=<X<;>@|<++<X@=@+<k<

}@=~jZX'=j~j'*V=V'X~j'*V=<=q&@*
~||

/+@;0

 `} `|| ~} `   ||}` }~ ^~ }}

 `^} ^`> ||  }~K `~ }`  |~ _` `^}~ 
_} `}` `~ ` ``> \} \^`Y^}^~| \^_`Y{^|}^~ 
`` ` |`|} ~^}   _} ~} `} `~}` } 
 ^}||`> ` ` _} } }}~  }~ |  ~|| -
} `}`> ||| } ~ } ^` }  ~} ~}`
|~| `}~  ` ||   ~ ~} _}}` `}}
}~``^|~_`^}`^^}`}`>}```
K^~}||}}`^`}~|~_``
`}~``>

29