You are on page 1of 8

SPE 105155

Wellbore Stability Analysis in UBD Wells of Iranian Fields


S. Salehi, SPE, and G. Hareland, SPE, U. of Calgary, and K. Khademi Dehkordi, NISOC Co.

Copyright 2007, Society of Petroleum Engineers concentration is produced around the well. If there is no hydrostatic
This paper was prepared for presentation at the 15th SPE Middle East Oil & Gas Show and support pressure introduced into the borehole, failure in the
Conference held in Bahrain International Exhibition Centre, Kingdom of Bahrain, 11–14 March formation may take place. Therefore, maintaining equilibrium in the
2007.
field to prevent rock failure requires the use of a support pressure
This paper was selected for presentation by an SPE Program Committee following review of which is usually provided by the drilling fluid. In order to evaluate
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to the potential for wellbore stability a realistic constitutive model must
correction by the author(s). The material, as presented, does not necessarily reflect any be used to compute the stresses and strains around a borehole (1, 2).Out
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of of the numerous published models, the linear elastic model (LEM) is
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper the most common approach. This is due to its simplicity and less
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than required input parameters compared to other models. However, using
300 words; illustrations may not be copied. The abstract must contain conspicuous these simple models in some cases underpredicts the wellbore stable
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435. ECD. The LEM based models do not adequately explain the fact that,
in many cases the borehole remains stable even if the stress
Abstract concentration around the borehole exceeds the strength of the
formation.
Drilling underbalanced is often used to prevent formation damage, Alternatively, elastoplastic models offer the ability to asses the
avoid lost circulation, and increase rate of penetration. However, it is mechanical integrity of a borehole more realistically. Westergaard
also risky and may lead to wellbore collapse due to lack of positive (1940) (3) published one of the early works contributing to the
support provided by the hydrostratic wellbore fluid column. Hence, knowledge of stress distribution around a borehole, in which an
the application of underbalanced drilling (UBD) should be evaluated elastoplastic model was developed. Later works using elasto-plastic
thoroughly through the use of in-situ stresses and rock mechanical models have been published (e.g., Gnirk, 1972; Risnes and Bratli,
properties to estimate under what hydraulic drilling conditions the 1981; Mitchell et al., 1987; Anthony and Crook, 2002) (2). Table 1
wellbore is stable. shows a summery of the current wellbore stability models (4).
An elastoplastic model for assessing wellbore stability analysis in
This paper presents a mechanical wellbore stability analysis for two
two depleted carbonate fields are presented in this paper. Based on
depleted Iranian fields, named herein as field A and B. Depleted
stability analysis results compared with behavior of the drilling of
Iranian fractured carbonate fields are suffering from severe wellbore
two UBD horizontal wells in these depleted fields proved the
stability problems and lost circulation during overbalanced drilling
feasibility and accuracy of using an elastoplastic model to predict the
conditions. The application of UBD in these fields with a pressure
operational ECD windows. Although considering behavior of only
less than formation pore pressure brought on new wellbore stability
two wells is not sufficient to conclude that implementing this model
problems like risk of shear failure and collapse of borehole wall.
for all the carbonate fields will yield similar results, it is suggested
Using good geomechanical model description matching field
that the result of this study can be expanded for use on other fields
characteristics in conjunction with rock failure criteria in some cases
with similar characteristics.
may lead to a good prediction for avoiding wellbore stability
problems and choosing the optimum mud weight window. By
analyzing cores, log and triaxial rock mechanical data, an Fields Background
elastoplastic model combined with a finite-explicit code was used in
the wellbore stability analysis to estimatimate an optimum Equivalent 90 % of the discovered fields in Iran are in carbonate reservoirs
Circulating Density (ECD) for these fields. Compared to some actual putting Iran as one of the largest carbonate producers in the world (5).
field data it was observed that using an elastoplastic constitutive Field A and B are located in southern part of Iran. Geological studies
model would be sufficient to analyze mechanical wellbore stability in combined with the core samples taken from these fields, represented
these fields. a highly fractured limestone reservoirs with low formation pore
pressure gradients. Because of the confidential nature of the studies
Introduction undertaken in these fields the names are not reveiled.
Lost circulation is one of the most severe drilling problems in these
fields. So the determination of proper ECD is challenging.
Underground formations are subjected to a vertical compressive
stress caused by the weight of the overlying strata and horizontal
stresses due to the confining lateral restraints. Under the action of ECD as a controllable factor
these in situ stresses, prior to drilling a borehole, the rock mass is in a
state of equilibrium that will be destroyed by the excavation. When a Some factors which affect wellbore failure can be controlled while
borehole is drilled, the load carried by the removed rock is then taken others are impossible to control because of the intrinsic properties of
by the adjacent rock to re-establish equilibrium. As a result, a stress earth. The primary factors affecting wellbore stability are listed in
support pressure introduced into the borehole, failure in the Table 2(3). Equivalent Circulating Density (ECD) is the dominant
2 SPE 105155

controllable factor in applied wellbore stability analysis. The support Technical Approach
pressure offered by the static or dynamic fluid pressure during
drilling, simulating, working over or producing of a well, will Different wellbore stability analysis methods are presented in
determine the stress concentration present in the wellbore vicinity(6). literature. Mclean et al (1) used finite element methods to predict
This paper estimates the optimum ECD for preventing wellbore wellbore stability parameters. The use of numerical/analytical models
stability problems in two depleted carbonate fields. to predict the mechanical behavior of a wellbore requires a number of
input parameters to be defined or assessed. Here we have used a
Data Gathering finite explicit code named FLAC (7) for conducting wellbore stability
simulations. An elastoplastic Mohr-Coulomb failure criteria model is
The operator was considering two underbalanced horizontal wells in used for assessing state of instability with respect to different ECDs.
two depleted carbonate fields. The 6 ½–in. wellbores were horizontal Since wellbore orientation and ECD values are the only controllable
at an approximate depth of 8687 ft in field A and 6218 ft in field B. factors a few assumptions were made;
Pore pressure gradient was found from DST test analysis for both .
fields at the mentioned depths and reported as 0.19 psi/ft and 0.27 • Orientation. The vertical direction is assumed to be a
psi/ft in field A and B respectively. principle direction of stress. Because there are some
Usually the most important data for wellbore stability analysis is the uncertainties to direction of the principle horizontal stresses
rock mechanical data. Knowledge of this type of data in Iranian fields the worst case scenario was assumed where the horizontal
is limited, however for the mentioned fields in this paper some wells are parallel to the maximum horizontal stress
laboratory test were available. direction (8).
For field A, two laboratory methods were used for determining • ECD. This was the only controllable variable and the
elastic properties. optimum ECD; considering UBD condition was determined
• Static (Triaxial) in the field A and B to be less than 5.57Lb/Gal and 5.67
• Dynamic (Acoustic) Lb/Gal respectively.
The preferred method is to perform static triaxial tests on cores since
this method provides the most accurate and reliable data. A criterion based on size of yielded zone was used in analyzing the
As a general rule, the dynamic techniques are inaccurate because of risk of borehole instability. Since the yielded zone will be susceptible
poro-elastic influence on sonic wave propagation. Consequently all to spalling due to pressure surges during trips and mechanical erosion
dynamic test methods must be calibrated to provide reasonable by the drillstring, the larger this zone is the greater the likelihood that
estimates of static values needed for wellbore stability analysis. instability-related problems will occur (9). (Figure 2) A parameter
Typically with frequency used in laboratory for acoustic often used as a borehole instability risk indicator is the Normalized
measurement, only small scale features dominate the results. So the Yielded Zone Area (NYZA), which is the cross-sectional area of the
static young modulus is taken as the more representative value for yielded rock around the borehole divided by the area of the original
stability analysis. In general it is found that the static poisson‘s ratio borehole. Experience has indicated that the onset of borehole
usually underestimates the stress in the reservoir, for this reason the instability problems is often associated with NYZAs greater than 1.0,
dynamic poisson‘s ratio was used for analysis. although the critical value for this parameter undoubtly varies
For the second field (B) compressive and shear sonic transit time depending on the setting and other factors such as well inclination
were used for calculating elastic properties (Equation 1). Morales and hole cleaning capacity. More details on this can be found in
correlation was used to convert dynamic values to static. (Appendix references (10, 11, 12).
A). As shown in Figure 3, FLAC was used for determination of yielded
zone area. The magnitude of maximum displacement vector should
always be considered in the acceptable range. Figure 4 illustrates the
ν d = (V P 2 2 2 2
-2V S ) [2 (V P -V S )] (1) typical FLAC output of the displacement vectors.
More than hundred simulations with different ECD values were
conducted; Figure 5, 6 shows the trend of NYZA changes with
Other data were found from a nearby undepleted reservoir, and then different ECDs in field A & B.
adjusted to depleted conditions. Because of the uncertainties in As shown in the figures with increasing ECD, the NYZA will
magnitude of cohesions, three different sets of data were used in the decrease and we will have more stable wellbore. Typically, drilling
wellbore stability analysis. (Table 4) with a bottomhole pressure above than formation pore pressure will
The total overburden stress gradient was approximated by integration decrease the risk of borehole instability due to less yielding area of
of formation bulk density over depth. (Equation 2) the rock adjacent to the borehole.
The plots for both fields indicates that there is relatively small
D max

amount of yielding (NYZA less than 1.0 ) predicted for pressures less
σV = ρ b g dD (2) than reservoir pore pressures. For instance, in field A ECD ranges
0
from 5.30-5.57Lb/Gal considering the UBD condition would be
Due to poro-elastic effects in these fields, the reservoir depletion acceptable. In the second field with respect to different cohesions, the
will directly affect the horizontal stresses. Figure 1 shows the stress ECD range is presented in Table 5. In the case of choosing 870 Psi
changes due to poro-elastic effects. Horizontal stresses were for cohesion, there would be no ECD value considering NYZA be
measured by the analysis of fracture pressure in these fields prior to less than 1.0. Based on the wellbore stability simulations, adjusting
depletion and then corrected to poro-elastic effects. NYZA criteria to 1.2 as a critical value provided adequate hole
Knowledge of the minimum in-situ stress direction in these fields was cleaning be maintained. An ECD range of 5.06-5.30 Lb/Gal was
limited to poor quality data from oriented caliper logs which defined recommended for drilling the horizontal sections of the mentioned
the directions as approximately N40W in field A and S20N in field well in field A. This was 0.27-0.47Lb/Gal less than reservoir pore
B.The rock mechanical properties for both fields are summarized in pressure. This difference was enough to guarantee underbalanced
Table 3&4. drilling conditions.
For field B 1450 Psi was used for cohesion which resulted in a
recommended ECD range of 5.40-5.48 Lb/Gal.
SPE 105155 3

Actual Well Response VS = shear velocity, m/s

The reservoir section of 6-1/2in. wellbore in field A was actually νd = dynamic poisson’s ratio
drilled with an ECD of 5.20 Lb/Gal so the underbalanced condition
was kept most of the time. No sever borehole stability problems was σV = overburden stress gradient, Psi/ft
reported during drilling of this section .The daily drilling reports also
indicates a decrease of 550% in mud loss compared with the
σ H max = maximum horizontal stress gradient, Psi/ft
overbalanced condition. σ H min = minimum horizontal stress gradient, Psi/ft
It is not possible to state whether the hole suffered any failure, since
no caliper were run in the 6-1/2 in. hole. No caliper or core data has φp = friction angle at peak strength, degree
currently been made available to assess the wellbore stability
condition of the well drilled in field B, but the well was drilled ρb = bulk density, kg/m
3

without any major complication as in Field A.


Acknowledgments
Conclusions and Recommendations
The authors would like to thank Mr. P. McLellan for his help on
1. Severe lost circulation during overbalanced drilling condition is some technical issues used in this paper and NISOC (National Iranian
reported during drilling depleted carbonate fields in Iran, for this South Oil Company) for their permission to publish the results
reason using the underbalanced drilling technique with proper presented in this paper.
wellbore stability analysis is recommended for drilling in these fields.
References
2. An elastoplastic model combined with a finite explicite code was
used for mechanical wellbore stability analysis of underbalanced 1. McLean, M.R. and Addis, M.A. (1990).Wellbore stability
drilling technique in two depleted Iranian fields. Based on the results analysis: A review of current methods of analysis and their
and compared with field data using elastoplastic models gives good field application. IADC/SPE Drilling Conf., Houston, TX,
predictions for wellbore stability in these fields. Feb., pp.261-274. IADC/SPE 1994.

3. A criterion based on size of yielded zone or NYZA (Normalized 2. Al-Ajmi, Adel. (2006).Wellbore stability analysis based on
Yielded Zone Area), was used to assess stability condition. Based on a new true-triaxial failure criterion. Ph.D. thesis.
the simulation results and for keeping UBD condition in most of the
time the critical value of NYZA was adjusted to 1.20 instead of unity 3. Westergaard, H.M., 1940 .Plastic state of stress around a
in Iranian carbonate fields. Choosing this critical value an ECD of deep well. J Boston Soc Civil Eng, 27, 1-5.
5.06-5.30 Lb/Gal was proposed for drilling the well in field A.
Compared with actual field data no wellbore stability problem was 4. Chen, Guizhong. (2001). A study of wellbore stability in
encountered during drilling of this well. However, more research and shales including poroelastic, chemical and thermal effects.
comparisons with field data is necessary to define proper NYZA PhD thesis, University of Texas at Austin.
critical value for stability assessment.
5. Abdollahi, J., Carlsen, I.M., Mjaaland, S., Skalle, P., Rafiei,
4. The Morales correlation was used for the field B to convert the A., Zarei S., (2004). Underbalanced drilling as a tool for
dynamic values to static. This correlation is for sandstone reservoirs optimized drilling and completion contingency in fracture
therefore the converted static properties are not necessarily correct to carbonate reservoirs. SPE/IADC 91579.
use in carbonate reservoirs as in the case of field B. Due to
uncertainties in rock mechanical properties presented in this paper, 6. Mclellan, P.J., Assessing the risk of wellbore instability in
core data is will be required to verify the properties used for field B. inclined and horizontal wells; Journal of Canadian
Petroleum Technology, Vol.35, No.5, pp.21-32, May 1996.
5. For complete wellbore stability analysis in these fields, a
combination of both mechanical and chemical effects should be 7. FLAC, Student License, Itasca Consulting Group, Inc.
considered. However, only the mechanical effects are considered in USA, 2006.
this paper.
8. Parra, J.G., Celis, E., and De Gennaro S., (2003). Wellbore
stability simulations for underbalanced drilling operations
Nomenclature in highly depleted reservoirs. SPE 83637.

C p = cohesion at peak strength, Psi 9. Hawkes, C. D., McLellan P.J. (1996). Modeling of yielded
zone enlargement around a wellbore. Proceedings of 2nd
Cr = cohesion at residual strength, Psi North American Rock Mechanics Symposium, Montreal,
D = depth, m Quebec, Canada.
DST = drill stem test
E = young’s modulus, Psi 10. McLellan, P.J. and Wang, Y., Predicting the effects of pore
ECD = equivalent circulating density, Lb/Gal pressure penetration on the extent of wellbore instability:
2 application of a versatile poro-elastoplastic model; paper
g = gravitational acceleration, 9.8 m/s SPE 28053, presented at SPE/ISRM Eurock’94, Delf, The
NYZA = Normalized Yielded Zone Area Netherlands August 29-31, 1994.

VP = compressional velocity, m/s 11. Hawkes, C.D. and McLellan, P.J., A new model for
predicting time-dependant failure of shales: Theory and
4 SPE 105155

Application; Petroleum Society paper 97-131 , presented at Table 1: Current wellbore stability models
the 48th annual technical meeting of the petroleum society , Reference Model Type Special Features
Calgary, AB, June 8-11 ,1997.
Bradley (1979a, Linear Elasticity Zero tensile strength
12. McLellan, P.J., Hawkes, C.D. and R.S. Read. (2000). Sand 1979b)
production prediction for horizontal wells in gas storage Fuh et al. (1988) No chemical effect
reservoirs. SPE/PS-CIM 65510. Aadnoy and No fluid diffusion Zero tensile strength
Chenevert (1987)
McLean and Addis No thermal effect
(1990a, b)
Appendix A Zhou et al. (1996)
Truncated Desai’s
Morales Correlation yield function

Esta = (-2.21*PIGN+0.963)*Edyn Santarelli (1987) Stress dependent


linear elasticity
Esta=Static Young’s Modulus
Edyn=Dynamic Young’s Modulus Detournay and Cheng Linear poroelasticity Vertical well
PIGN=Log Porosity (1988) Undrained condition

Yew et al. (1990) Moisture adsorption


Wang (1992) Chemical effect and Variable Young’s
moisture adsorption modulus
Yew and Liu (1992) Linear poroelasticity,
no chemical effect
Mody and Hale Chemical effect
(1993a, b) Stress on the wellbore
wall
Sherwood (1993) Chemical effect Chemical potential of
different components
Wang et al. (1994) Chemical effect Shale properties vary
with water content
Cui et al. (1997) No chemical effect Solution in Laplace
Cui et al. (1999) Time dependency domain, superposition
Time dependency technique

Abousleiman et al. Poroviscoelasticity


(1996)

Addis and Wu (1993) Nonlinear elasticity

McLean (1989) Elasto-Plasticity


McLellan and Wang Elasto-Plasticity
(1994) Plasticity
Ewy (1991, 1993)

Detournay (1995) Coupled thermo- Drained and undrained


Wang et al. (1996) hydro-poroelasticity conditions
Li et al. (1998) Thermoporoelasticity Conductive heat flow
Choi and Tan (1998) Thermoporoelasticity
Numerical validation

Wang and Dusseault Thermoporoplastic


(1995)
SPE 105155 5

Table 2: Controllable Factors and Uncontrollable Factors Table 5: ECD range for UBD condition regarding different cohesion
values
Cohesion at peak strength(Cp,Psi) ECD Range (Lb/Gal)
Controllable Factors Uncontrollable Factors Considering UBD condition
2175 5.4-5.7
Equivalent Circulating In-situ stresses 1450 5.67-5.7
Density(ECD) Rock lithology 870 N/A
Drilling fluid type Pore fluid chemistry
Drilling fluid chemistry Rock porosity, original
Well orientation, direction permeability, compressibility
relative to stress field Initial rock temperature
Mud temperature Rock strength
Borehole size Rock mechanical properties
Drill pipe size Initial pore pressure
Circulating rate Natural fractures
Open hole time Rock thermal properties
Drilling operations (tripping, Geothermal gradient
drilling,casing, cementing)

Table 3: Input parameters for wellbore stability analysis in field A

Poisson’s Ratio 0.33

6 3.31
Young’s Modulus(E,10 Psi)
Friction Angle at Peak Strength ( φ p ) 41.50

Cohesion at Peak Strength(C p ,Psi) 1850

Cohesion at Residual Strength (C r ,Psi) 540

Overburden Stress Gradient( σ v ,Psi/ft) 1.20

Maximum Horizontal Stress Gradient( σ H max , Psi/ft) 0.81


Figure 1: Poro-elastic effects.
Minimum Horizontal Stress Gradient ( σ H min , Psi/ft) 0.72

Table 4 : Input parameters for wellbore stability analysis in field B

Poisson’s Ratio 0.30


6 0.66
Young’s Modulus(10 Psi)

Friction Angle at Peak Strength ( φ p ) 30,35,40

Cohesion at Peak Strength(C p ,Psi) 870,1450,


2175
Cohesion at Residual Strength (C r ,Psi) 217,480,
620
Overburden Stress Gradient( σ v ,Psi/ft) 1

Maximum Horizontal Stress Gradient( σ H max , Psi/ft) 0.69

Minimum Horizontal Stress Gradient ( σ H min , Psi/ft) 0.58


6 SPE 105155

Figure 2: Development of yielded zone.

Figure 4: Maximum displacement vectors.


SPE 105155 7

Figure 3: Predicting yielded zone in FLAC.

Figure 5: NYZA versus ECD in field A.


8 SPE 105155

Figure 6: NYZA versus ECD in field B.