Evaluation of The Relative Importance of Parameters Influencing Perforation Cleanup

Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

SPE 86538

Evaluation of the Relative Importance of Parameters Influencing Perforation Cleanup


Russell L. Detwiler, SPE, and Joseph P. Morris, Lawrence Livermore National Laboratory; C. Ozgen Karacan, SPE, Phillip
M. Halleck, SPE, Penn State University; and John Hardesty, SPE, Jet Research Center

This paper was prepared for presentation at the SPE International Symposium and Exhibition
on Formation Damage Control held in Lafayette, Louisiana, U.S.A., 18–20 February 2004.
(underbalanced) to promote a transient pressure surge and
high velocity flow through the perforation and adjacent region
This paper was selected for presentation by an SPE Program Committee following review of
information contained in a proposal submitted by the author(s). Contents of the paper, as of the formation. The underbalance-induced surge results in
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
partial removal of debris and damaged rock from the
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at perforation, leaving an open tunnel surrounded by a zone of
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper damaged rock. Experimental results suggest that this removal
for commercial purposes without the written consent of the Society of Petroleum Engineers is of debris and damaged rock occurs on a time-scale
prohibited. Permission to reproduce in print is restricted to a proposal of not more than 300
words; illustrations may not be copied. The proposal must contain conspicuous of milliseconds.3
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. A variety of models have been proposed to simulate
different stages of the perforation process and provide
estimates of the post perforation permeability of the damaged
Abstract
Completion of cased and cemented wells by shaped-charge zone. Pearson and Zazovsky4 developed a model of the
perforation results in damage to the formation, which can transport of sand grains through the perforation tunnel to the
significantly reduce well productivity. Typically, well bore to demonstrate the minimum underbalance required
underbalanced conditions are imposed during perforation in an to remove debris from the perforation tunnel. Arora and
effort to remove damaged rock and shaped-charge debris from Sharma4 modeled the migration of fines within the formation
the perforation tunnel. Immediately after the shaped-charge jet and calculated their influence on the post-perforation
penetrates the formation, there is a transient surge of fluid permeability of the damaged zone. Additional efforts have
from the formation through the perforation and into the well aimed at understanding the three-dimensional flow geometry
bore. Experimental evidence suggests that it is this transient near perforations under downhole conditions.,5,6. While these
pressure surge that leads to the removal of damaged rock and models address different aspects of the perforating process,
charge debris leaving an open perforation tunnel. they all assume an initial perforation tunnel geometry and
We have developed a two-stage computational model to permeability distribution within the surrounding rock. We
simulate the perforation process and subsequent pressure surge present a model that explicitly calculates both the permeability
and debris removal. The first stage of the model couples a distribution and the final post surge geometry of the
hydrocode with a model of stress-induced permeability perforation tunnel in Berea sandstone. Our model consists of
evolution to calculate damage to the formation and the two sequential stages: 1) jet penetration of the casing and rock
resulting permeability field. The second stage simulates the formation and 2) underbalanced-induced pressure surge and
non-Darcy, transient fluid flow from the formation and debris removal. We compare measured perforation tunnel
removes damaged rock and charge debris from the perforation geometries and core-flow efficiencies (CFEs) for a series of
tunnel. We compare the model to a series of API RP19B API RP19B section 4 flow tests in Berea sandstone cores to
section 4 flow tests and explore the influence of fluid viscosity simulated estimates for the same underbalance conditions.
and rock strength on the final perforation geometry
and permeability. Model Description
We have developed a two-stage perforation model to
investigate the relative importance of the parameters that
Introduction
Shaped-charge perforation is often used to complete cased influence perforation tunnel geometry and the final
wells. Detonation of a shaped-charge creates a high-velocity permeability distribution surrounding the perforation tunnel.
(~7000 m/s) jet of metal particles that penetrates the well The first stage couples a hydrocode used to simulate
casing and the adjacent formation. The jet leaves a path of mechanical deformation of the rock caused by the shaped-
debris consisting of crushed rock and metallic particles while charge jet with a model of stress-induced alteration of the rock
generating a shock wave that propagates outward through the permeability. This model thus provides a detailed map of the
formation. The shock wave damages rock surrounding the permeability distribution in the formation prior to the onset of
perforation, resulting in a region with altered permeability.1,2 flow from the formation that is based upon independently
Typically, prior to shaped-charge detonation, the wellbore measured rock properties. The second stage solves for the
pressure is reduced to below the pore pressure in the reservoir transient flow of fluid from the formation and the resulting
2 SPE 86538

removal of damaged rock from the perforation tunnel. Because immediately after perforation in which inertial forces likely
the time scale of the jet penetration and shock wave play a significant role.12,13. We solve for the transient pressure
propagation is considerably smaller than the time scale for distribution within the formation using
equilibration of post-perforation fluid pressures, these
φβ f µ ∂P
processes are simulated sequentially. Furthermore, the models ∇ 2P = ……………………………………….(1)
assume radial symmetry and thus solve the problems in two k ∂t
dimensions (Fig. 1). which assumes a slightly compressible fluid where βf is the
Hydrocodes used to simulate shaped-charge perforations fluid compressibility µ is the dynamic viscosity, φ is the
can provide estimates of the extent of the damaged zone, but porosity and k is the local permeability. Because inertial terms
do not typically provide estimates of the post-perforation may be important, especially at early times, we use the
permeability distribution within the damaged formation. Forchheimer equation,
Morris et al.7 presented a model for predicting the evolution of
the porosity and permeability of Berea sandstone undergoing 1 ⎛µ r ⎞r
−∇P = ⎜ + Β u ⎟u ,.……………………………….(2)
low-strain-rate deformation. We modify their model to ρ⎝k ⎠
accommodate the high-strain-rate conditions encountered
during jet perforating and incorporate this modified porosity- to represent P in solving (1) for the pressure distribution,
permeability model into a hydrocode used to simulate where ρ is the fluid density and B=1.1116x10-12/k1.55 is the
jet penetration.8 empirically derived coefficient of inertial resistance,13 which
Morris et al. used an elastic-plastic material model.8,9 to accounts for the increasing importance of inertial forces at
determine the stress-strain response of Berea sandstone and higher fluid velocities. At low fluid velocities, (2) reduces to
quantified the total porosity as the sum of incipient porosity Darcy’s equation. We use a centered-in-space, backward-in-
and dilatant porosity. They assumed that changes in time finite difference scheme to solve (1) during the post-
permeability result only from alterations in the incipient perforation surge. Adaptive time stepping ensures small
porosity (see Appendix A for details). The model changes in P and k between time steps.
demonstrated good agreement with low-strain-rate Removal of damaged rock from the tunnel walls occurs
experimental measurements presented by Zhu and Wong.10 during the initial pressure surge and is determined locally by
To extend this approach to the high-strain-rate conditions the magnitude of the pressure gradient within each grid block.
that occur during jet perforating we assume the fluid does not The threshold for removing damaged rock is a combination of
have time to flow in response to porosity changes. Thus, the the tensile and shear strength of the damaged rock. We assume
fluid is confined within individual pores and local fluid that the tensile strength of the rock surrounding the perforation
pressure responds to changes in pore volume. This limit may is considerably smaller than the shear strength. Thus, if in any
be thought of as corresponding to an undrained experiment. time-step the local pressure gradient exceeds the
An increase in pore pressure results in a reduction in the critical value,4
effective stress and consequently the apparent confinement σt
∇Pcr = ,………………………………………………(3)
and strength of the rock is reduced. This high strain rate 2a
approach is presented in Morris et al.11 we assume that the damaged rock in that grid block is
Morris et al.7 assumed that although the dilatant porosity loosened from the wall and eventually swept from the
may be connected it does not make a substantial contribution perforation tunnel. This assumes that once material is loosened
to flow. This is consistent with the dilatant porosity having from the tunnel wall, there is adequate flow velocity in the
poor connectivity or higher surface area to volume ratio. perforation tunnel to carry the debris to the wellbore, or if not,
Using this approximation, the experimentally measured that the permeability of the loose debris is sufficiently high
porosity is the sum of the surviving incipient porosity and that it does not significantly impede flow through the
induced dilatant porosity. However, only the incipient porosity perforation tunnel. σt is the tensile strength of the rock and a is
contributes to flow, resulting in a model for permeability that a characteristic grain diameter of the rock along the tunnel
is related to incipient porosity, effective stress history, and a wall, which we estimated to be 0.1 mm for the simulations
damage variable (see Eq. A12). Due to the lack of presented in this paper.
experimental data regarding stress-induced permeability The plastic strain experienced by a material element during
alteration in anisotropic media, the current model assumes an the hydrocode simulation provides an indication of the
isotropic permeability field. damage sustained by that element. We relate the local tensile
Fig. 2 shows an example of the permeability field that strength of the rock, σt, to the plastic strain (εp) that the rock
results from simulating jet penetration of a core of Berea was subjected to during jet penetration,
sandstone. The permeability of the debris in the damaged zone
⎛ 0.01 ⎞
is reduced by several orders of magnitude. However, this σ t = σ o min⎜⎜ ,1.0⎟⎟ ,……………………………….(4)
material has also been considerably weakened due to grain ⎝ εp ⎠
fragmentation, and will likely be removed from the perforation
tunnel during the post-perforation pressure surge. where σo is a reference value for undamaged rock. We use a
Upon penetration of the formation, underbalance causes a value of 0.01 GPa for the simulations in Berea sandstone
transient expansion of the higher-pressure fluid in the presented in this paper. Once ∇Pcr is exceeded, solids are
formation and flow towards the opening created in the well removed from the element over a time scale of ∆x v and k
casing by the jet. This results in large pressure gradients
SPE 86538 3

increases linearly from ko to kmax. Fig. 3 shows the evolution results in larger tunnel radii, particularly near the well bore.
of permeability and pressures within the simulated core during The radii of the simulated perforation exhibit a stronger
the post-perforation pressure surge. dependence on underbalance than the experiments. This may
Initially the fluid in the reservoir is at a uniform pressure in part be due to the form of Eq. 4, which has not been
and the perforation tunnel is completely filled with damaged independently verified. In addition, the radii of the simulated
rock and charge debris. Flow is then initiated by setting the tunnels vary more smoothly than the experiments, which may
well bore pressure to the desired underbalance (750 psi for this be due to layering in the rocks resulting in local variability in
example). Debris is removed axially along the length of the σt and k that are not represented in the model. Though there
tunnel until local pressure gradients are insufficient to are discrepancies between the simulations and experiments, it
overcome the required yield stress. At steady-state there is a is encouraging that a model largely based upon independently
slightly wider damaged zone near the well bore which results measured rock properties can provide reasonable predictions
in substantially larger pressure gradients near the tunnel of physical experiments.
entrance. Conversely, the tip half of the tunnel has a thinner To further evaluate the capability of the simulations to
damaged zone but smaller pressure gradients. predict experimental observations, we compare the CFEs
We calculated local mass fluxes to investigate the calculated for the experiments to the simulated CFEs (Fig. 6).
influence of axial variability in the extent of the damaged zone CFE is calculated as the ratio of the permeability of the
and the magnitude of pressure gradients on flow to the perforated core to the permeability of a core with the same
perforation tunnel. Fig. 4 shows steady-state mass flux in the perforation tunnel geometry with no damage (i.e. k=ko). For
region surrounding the perforation tunnel for perforations at medium to high underbalances, the simulations predict CFE
750-psi and 3000-psi underbalance. For the 750-psi case, the values reasonably well, but at low underbalance, the simulated
large damaged zone observed in Fig. 3 remains intact and CFEs significantly underestimate experimental values. This
impedes flow into the tunnel near the entrance despite larger may result from the nature of debris filling the tunnel. X-ray
pressure gradients in this region relative to the tip. Conversely, CT scans of the tunnels show metal charge debris plugging the
the additional removal of damaged rock from near the tip of the perforation tunnel in each of the experiments15. The
entrance at 3000-psi underbalance results in larger mass fluxes remaining volume of metal decreases with increasing
near the tunnel entrance than near the perforation tip. This underbalance, however, for the 750-psi underbalance core,
change in flow regime from low to high underbalance is in there was an annulus surrounding the metal plug through
part due to the pressure boundary conditions applied during which fluid can enter the tunnel. This flow conduit, which was
the simulations. As with API RP19B section 4 flow tests, we not observed at higher underbalances, may be the cause of the
prescribe uniform constant pressure boundary conditions large measured CFE for the 750-psi experiment. While the
along the length of the core. Thus, underbalances that are hydrocode tracks different material during the perforation
sufficiently high to remove the damaged rock surrounding the process, the permeability and rock strength models do not
tunnel entrance result in larger pressure gradients and more differentiate between damaged rock and metal charge debris.
flow in this region. However, under downhole conditions, the Thus, the simulations cannot replicate the variable influence of
presence of multiple perforations and the three-dimensional the metal debris at the tip of the perforation tunnel.
geometry of the flow field will result in a pressure field that Enhancement of the computational model to incorporate
favors flow to the perforation tip.5,6 This suggests that models for both materials (damaged rock and metallic charge
enhanced CFEs observed at high underbalance may debris) may help to gain new insights on the influence on
exaggerate the actual productivity of perforations under CFEs of metal plugging the perforation tip.
field conditions. To explore the role of fluid and rock properties on post-
perforation tunnel geometry and CFE we conducted a series of
Comparison of Computational and simulations in which we varied the tensile strength of
Experimental Results undamaged rock and the fluid viscosity. Fig. 7 shows CFEs
We test the two stages of our model through comparisons with measured for simulations in which we doubled and halved the
results from a series of API RP19B section 4 flow tests on values of σo and µ used for the base simulations presented in
cores of Berea sandstone at underbalances ranging from 750 Fig.6. Values of σo and the details of the model chosen to
psi to 3000 psi. The 0.178 x 0.394 m (7 x 15.5 in) cores were represent rock strength surrounding the perforation tunnel
cut perpendicular to the bedding planes resulting in axial strongly influence calculated CFEs. At lower underbalances,
permeabilities of ~125 mD and cross diameter permeabilities doubling σo results in CFEs that are less than half of those
of ~55 mD. These tests provide a bulk measurement of the estimated for the base simulations. However, changes in
change in permeability of the core resulting from perforation. viscosity, which influence the rate of dissipation of the
After completion of the flow tests, the cores were scanned underbalance-induced pressure gradients have a negligible
using high-resolution x-ray CT techniques.15. This provided influence on calculated CFEs.
detailed three-dimensional images of the geometry of the
perforation tunnels and the locations of any remaining charge Conclusions
debris within the perforation tunnels. Fig. 5 compares open • Creation of perforation tunnels and the subsequent clean
tunnel radius plotted against tunnel length for the experiments up by underbalance-induced pressure surge in sandstone
and simulations. The simulated perforation tunnels exhibit can be predicted reasonably well with a sequential model
similar characteristics to the experimentally measured tunnels: that first simulates jet penetration and permeability
they decrease in radius towards the tip and larger underbalance
4 SPE 86538

alteration followed by simulation of the transient non- 11. Morris, J.P., Lomov, I.N., and Glenn, L.A.: “Simulating
Darcy pressure surge. Perforation Permeability Damage and Cleanup”, SPE
• Simulations suggest that at low underbalance, most of the 71316, 2002.
12. Halleck, P.M.: “Advances in understanding perforator penetration
flow to perforation tunnels occurs toward the tip end of
and flow performance”, SPE 27981 presented at the
the tunnel, whereas at high underbalance, flow to the University of Tulsa Centennial Petroleum Engineering
perforation tunnel is greater near the tunnel entrance. Symposium, Tulsa, Oklahoma, August 29-31, 1994.
Thus, high CFEs measured at large underbalance may 13. Arora, D.S. and Sharma, M.M.: “The nature of the compacted
overestimate the increased productivity of a perforation zone around perforation tunnels”, SPE 58720 presented at the
under downhole conditions. International Symposium on Formation Damage Control,
• Small changes in tunnel radius induced by variability in Lafayette, Louisiana, Feb. 23-24, 2000.
rock strength may lead to small alterations in the 14. Halleck, P.M., Karacan, C.O., Hardesty, J., and Detwiler, R.:
“Changes in perforation-induced formation damage with
thickness of the damaged, zone and thus, significant
degree of underbalance: Comparison of sandstone and
variability in CFE estimates. limestone formations”, SPE 86541 presented at SPE
International Symposium and Exhibition on Formation
Acknowledgments Damage Control, Lafayette, Louisiana, Feb. 18-20, 2004.
This work was performed under the auspices of the U. S.
Department of Energy by the University of California, Appendix A
Lawrence Livermore National Laboratory under Contract No. Here we present a summary of the relevant details of the
W-7405-Eng-48. The authors thank the management of porosity-permeability model under quasistatic deformation
Halliburton Energy Services, Inc. and Jet Research Center presented by Morris et al.7,11 The evolution of the incipient
permission to publish this paper. porosity, which is used to calculate permeability, is
described as
References
1. Papamichos, E., Vardoulakis, I., and Ouadfel, H.: Permeability
⎧Φ − σ Φ − Φ + CP
⎪ I
φI = ⎨
o eff I * ( (
* ))
P* for σ eff < P* ,
(A1)
reduction due to grain crushing around a perforation. Int. J. ⎪⎩Φ * + Cσ eff otherwise,
Rock Mech. Sci. & Geomech. Abstr., 30(7):1223–1229, 1993.
2. Asadi, M. and Preston, F.W.: “Characterization of the crushed- where,
zone formed during jet perforation by qualitative scanning
σ eff = (σ 1 + σ 2 + σ 3 ) 3 − Pp = Pm − Pp ……………….(A2)
electron microscopy and quantitative image analysis”, SPE
22812 presented at the SPE Annual Technical Conference and The material constants, ΦI, C, P*, and Φ* along with other
Exhibition, Dallas, Texas, October 6-9, 1991.
values specific to Berea sandstone are provided by Morris et
3. Bartusiak, R., Behrmann, L.A., and Halleck, P.M.: “Experimental
investigation of surge flow velocity and volume needed to al.7 Large hydrostatic loads and non-hydrostatic loads can lead
obtain perforation clean up”, J. of Pet. Sci. Eng., 17(1997): to grain crushing leading to additional reduction in porosity
19-28. and permeability. To account for formation damage, Morris et
4..Pearson, J.R.A. and Zazovsky, A.F.: “A model for the transport of al.7 devised an adjustment to (A1),
sand grains from a perforation during underbalance surge”,
SPE 38634 presented at SPE Annual Technical Conference
φ IS =
( ) ………………………………………..(A3)
φ Io σ eff
max

and Exhibition, San Antonio, Texas, Oct. 5-8, 1997.


1.0 + D
5. Tariq, S.M.: “Evaluation of flow characteristics of perforations
including nonlinear effects with the finite-element method”, where D evolves according to
SPE Production Engineering (May 1987) 104.
6. Anash, J., Proett, M.A., and Soliman, M.Y.: ”Advances in well (
Dś = c1εśp min σ eff , Ps* + ) ………………..(A4)
completion design: A new 3D finite-element wellbore inflow
model for optimizing performance of perforated
( )( )
c2 H σ eff − Ph1 H Ph2 − σ eff −νś
completions”, SPE 73760 presented at SPE International where c1 and c2 are constants, εp is the plastic strain, and v is
Symposium and Exhibition on Formation Damage Control, the trace of the velocity tensor (positive indicates expansion).
Lafayette, Louisiana, Feb. 20-21, 2002. H is the Heaviside function and 〈x〉=xH(x) . The first source
_ _

7. Morris, J. P.; Lomov, I. N.; Glenn, L. A.:: “A constitutive model term increases damage in response to plastic strain. The
for stress-induced permeability and porosity evolution of
second term causes damage in response to reductions in
Berea sandstone,” J. Geophys. Res., Vol. 108, No. B10, 2485,
10.1029/2001JB000463, 18 October 2003 volume at high hydrostatic loads and Ph1 and Ph2 are the lower
8. Lomov, I.N., Hiltl, M., Vorobiev, O.Y., and Glenn, L.A.. and upper bounds on the effective mean stress for which this
“Dynamic behavior of Berea sandstone for dry and water- damage term is active. The parameters in Table A1 were
saturated conditions”, Int. J. Impact Eng., 26(1-10):465-474, found to give good agreement with experiment.7
2001. We extend this approach to dynamic problems by
9. Rubin, M.B., Vorobiev, O.Y., and Glenn, L.A.: “Mechanical representing the incipient porosity ( φ ID ) as the sum of pore
and numerical modeling of a porous elastic-viscoplastic
material with tensile failure”, Int. J. Solis and Structures, 37, volume that would open under static conditions ( φ IS ) and pore
1841-1871, 2000. volume that is forced open by the expanding fluid. This
10. Zhu, W. and Wong, T.-f.: “The transition from brittle faulting to results in
cataclastic flow: Permeability evolution”, J. Geophys. Res.,
102(B2):3027–3041, 1997.
SPE 86538 5

σ eff
(Φ ))
Nomenclature
φ ID = Φ I −
σ max I (
− φ IS σ eff
max
,D …………………..(A5) a = characteristic grain diameter (m)
eff
B = Forchheimer coefficient (m-1)
where σeffmax is the maximum effective stress over the history c1, c2 = damage evolution constants
of the material element. The form of (A5) ensures continuity C = incipient porosity reduction rate (Pa-1)
between the fluid porosity during loading and unloading and D = damage factor
causes the fluid porosity to return to the reference value when Dmax = maximum damage factor
the effective mean stress is zero. The fluid porosity is then E = incipient porosity increase rate (Pa-1)
given by: F = incipient porosity intercept
J = ratio of specific volumes
k = permeability (m2)
⎧⎪φ S , for σ > σ max k* = initial permeability (m2)
φI = ⎨ D
I eff eff
………………………..…(A6) ko = initial post-perforation permeability (m2)
⎪⎩φ I , for σ eff ≤ σ eff
max
kmax = maximum permeability after clean up (m2)
K1 = permeability-porosity exponent
For the dynamic case, pore pressure is not prescribed P = pressure (Pa)
externally and we introduce an extra equation to close the ∇Pcr = critical pressure gradient (Pa/m)
system and solve for the pore pressure: Pp = pore pressure (Pa)
Pm = mean stress (Pa)
1 SΦ I Ph1 = minimum pressure for hydrostatic damage (Pa)
Pp = −1 …………………………………...(A7)
βf φI J
Ph2 = maximum pressure for hydrostatic damage (Pa)
where Kf is the fluid bulk modulus, J is the ratio of current and P* = critical effective pressure (Pa)
reference specific volumes of the mixture, and S is a saturation Ps* = maximum pressure-enhanced strain damage (Pa)
parameter (assumed constant during dynamic experiments). r = radial dimension (m)
Equation (A7) was derived assuming that the pore fluid is S = saturation parameter
confined and obeys a linear equation of state. Eq. A7 refers to
t = time (s)
the fluid porosity, so Pp must be eliminated from (A7) and u = mass flux (kg/m2/s)
(A6) to obtain the fluid porosity. Regardless of which portion v = fluid velocity (m/s)
of Eq. A6 applies, φI is linear in σeff. The intersection between x = axial dimension (m)
(A7) and a straight line: ∆x = grid spacing (m)
φ I = Eσ eff + F …………………………………..….…(A8)
Greek Symbols
⎛ ⎞ βf = fluid compressibility (Pa-1)
⎝ ⎠ ( )
φ I = ⎜φ * + φ *2 − 4FSΦ I / β f J ⎟ 2 …………………..(A9) εp = plastic strain
φ = porosity
where
φI = incipient porosity
φ * = F + E(Pm +1 β f ) ……………………………….(A10) φ Io = undamaged incipient porosity
Back substitution into (A6) yields the effective mean stress. φ IS = incipient porosity for low-strain-rate
Morris et al.8 obtained good agreement with (quasi-static) deformation
experimentally measured permeabilities using φ ID = incipient porosity for high-strain-rate

{ ( )
k = k * exp K1φ1 σ eff ,D − D min D,D max ( )}…………(A11) (dynamic) deformation
Φ* = incipient porosity intercept
ΦI = reference incipient porosity
µ = dynamic viscosity (Pa-s)
θ = angular dimension
ρ = fluid density (kg/m3)
σ1, σ2, σ3 = priniciple stresses (Pa)
σeff = effective stress (Pa)
σ eff
max
= maximum effective stress (Pa)
σt = tensile yield stress (Pa)
σo = tensile yield stress of undamaged rock (Pa)
ν = trace of the deformation tensor
6 SPE 86538

r
θ . .

Fig. 2.― Log-permeability distribution after jet penetration


and prior to onset of post-perforation pressure surge in a
9-cm diameter, 19-cm long core. The color scale (blue-
Fig. 1. ― Schematic of model domain. The model yellow-red) represents increasing permeability
assumes radial symmetry, so simulations are carried
out in the r-x plane.

Fig. 3. ― Evolution of permeability in perforation tunnel (left) and pressure distribution (right) during post-perforation
transient surge from reservoir to well bore (left edge of domain). Gray represents open perforation tunnel.

Fig. 4.― Mass flux around perforation tunnel at steady state for 750-psi
underbalance (bottom) and 3000-psi underbalance (top). The color scale
(red-yellow-blue) represents increasing mass flux.
SPE 86538 7

2.5
Underbalance (psi)
2 750
1500
Tunnel radius (cm)
3000
1.5

0.5

0
0 5 10 15 20
Tunnel length (cm)
Fig. 5. ― Comparison of measured (solid lines) and simulated (data points) tunnel
diameter along the tunnel length.

1.5
1.5 base simulation
Experiments σ o = 0.5 σ o,base
Simulations σ o = 2 σ o,base
µ = 0.5 µ base /2
1 µ = 2 µ base
1
CFE
CFE

0.5
0.5

0 0
0 1000 2000 3000
0 1000 2000 3000
Underbalance (psi)
Underbalance (psi)

Fig. 7. ― Influence of fluid viscosity and rock


Fig. 6. ― Experimental and simulated CFEs for tensile strength on calculated CFEs.
different underbalances

You might also like