SPE 94252

Investigating Applicability of Vogels IPR for Fractured Wells

J.I. Rueda,* SPE, A. Zakharov, SPE, and J. Mach, SPE, YUKOS E&P

* Now with Pinnacle Technologies

Copyright 2005, Society of Petroleum Engineers Inc. correlation is now being used for different fields and better fits
This paper was prepared for presentation at the 2005 SPE Production Operations Symposium
the data than Vogel's correlation.
held in Oklahoma City, Oklahoma, U.S.A., 17 19 April 2005.

This paper was selected for presentation by an SPE Program Committee following review of Introduction
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
correction by the author(s). The material, as presented, does not necessarily reflect any In 1968 Vogel1 developed a correlation to estimate IPR curves
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at for two-phase flow. Vogels study was based on a numerical
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 simulation, assuming radial flow, initial reservoir conditions at
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 300
the bubble pressure, undamaged well and pseudo-steady state.
words; illustrations may not be copied. The abstract must contain conspicuous In his simulations, Vogel used 4 sets of PVT and relative
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. permeability data and showed that at those conditions AOF
decreases 1.8 times due to multi-phase effect. However,
Abstract Vogels correlation is often used in a wider range of
conditions than it was developed, including fractured wells,
It is a common practice in the oil industry that production reservoirs above bubble pressure and steady-state flow.
engineers use Vogel's correlation to correct the IPR curve Vogels deliverability curve is described by the following
below the bubble pressure for unfractured and fractured wells. equation:
However, there has not been a comprehensive investigation to
ensure if the Vogels correlation can be applied for fractured
wells.
q0
qo,max
= 1 0 .2
pw
pr 0 .8 ( )
pw 2
pr

This paper presents a new correlation to build IPR curves or (1)

kh
predict production performance below the bubble pressure for qo ,max = p r 1 .8
fractured wells. In order to investigate fractured well 141.2[ln (re / rw ) 3 / 4 + s ]
performance below bubble point, about 1,000 simulations runs
were performed using well-refined size grid for several sets of
Here qo ,max is the maximum possible flow rate (AOF), and qo
relative permeability curves and PVT data. The simulation
model has been validated against analytical solution. Those is the flow rate corresponding to wellbore flowing pressure pwf
runs cover a big practical range of fracture penetration 0.1 to and average reservoir pressure, pr.
1.0 and dimensionless fracture conductivities from 0.5 to 50.
For damaged or stimulated wells, Standing2 developed a
Steady state conditions were analyzed at this study. All the
modification to Vogels correlation which uses flow efficiency
mechanisms that cause the difference between fractured well
as a parameter:
and radial flow performance below the bubble pressure has

( ) 0.8 FE (1 )
been also well studied and will be presented in this paper.
qo p wf p wf 2
=1
= 1.8 FE 1 pb
2
pb (2)
It was found that Vogel's correlation underestimates fractured qoFE,max
well performance below bubble point. Vogel suggests a
correction of AOF by 45% meanwhile the simulation results
and new correlation show that the correction should be only Note, that this equation is applicable only for values
22%. Therefore, engineer could have an error of 43% using pwf Pr 1 FE(
1
)
to avoid reversing effect on the IPR curve.
Vogel for estimating AOF for a fractured well. Craft and Hawkins3 presented the eqn (3) to calculate the
Another finding of this study is that multiphase effect is absolute open flow rate.
dependent on fracture conductivity and almost independent on
qo ,max
fracture penetration. Higher conductivity fractures has bigger
=1
= 0.624 + 0.376 FE (3)
gas banks therefore they are affected by multi-phase effect to qoFE,max
greater extent than lower conductivity ones. The new
2 SPE 94252

Many engineers use Vogels equation, eqn 1, to correct IPR below pb For Unfractured Wells
production performance for fractured wells. It will be shown
that below that Vogel undestimates the production behavior The general IPR relationship for bottom flowing pressure
for fractured wells. Moreover, since a stimulated well can be below the pb has the form of eqn (7), as follows:
considered as a well with negative skin (or efficiency, FE
greater than 1), it is not uncommon that engineers apply eqn
(1), or eqns (2) and (3) to estimate the IPR curve for fractured pr 1 pb k ro
wells. As it will be shown below, this approach leads to q = 1 J D kh dp + dp , pwf < pb
overestimation of well performance. pb o Bo p wf B
o o
In this work, a simulation study has been conducted in order to
(7)
investigate the applicability of Vogels IPR to fractured wells.
The relative permeability curves and fluid properties that
The second term in the equation is, in fact, the reason for
Vogel used, were also used in our simulation study.
inefficient behavior of IPR curve below pb. Figures 1 and 2
Additionally, the relative permeability and fluid properties of a
below show the behavior of the mobility ratio below and
major field in western Siberia-Russia was added to our set of
above the bubble point. From this behavior it is clear that it is
data of our analysis. A big range of fracture penetrations and
necessary to divide the mobility ratio integral into two to
conductivities were considered in our study, simulating
cover the range of the mobility ratio, from the bottom flowing
constant pressure boundary conditions or steady state.
pressure (below the bubble point) to the average reservoir
pressure.
IPR above Pb For Unfractured Wells
While evaluating an efficiency of production enhancement, it Fractured Model Description
is very convenient to use IPR curves. To develop an analytical
A commercial black-oil numerical model was the one used for
solution for IPR above the bubble point, the following
study a square reservoir of constant height with a fractured
equation could be used
well producing in the center. Due to symmetry presented in
our model, only one quarter of the reservoir was simulated.
q = 1 J D
ko h
( pr pwf ) (4) For the simulation, an irregular Cartesian grid was used in
o Bo order to decrease simulation time and increase the accuracy of
calculations. The fracture has been modeled as a number of
where 1 - constant dependant on unit system cells with changed permeability and porosity. To simulate
constant outer boundary pressure, to be able to reach steady-
JD - dimensionless productivity index state, an infinite-conductivity big size oil reservoir connected
to the opposite site of the fracture was modeled.
The integral form of eqn (4) is as follows:
The fluid PVT properties and relative permeability curves that
pr k ro
q = 1 J D kh dp (5)
Vogel used for his research and an additional PVT and relative
p wf o Bo permeability set of data from a major western Siberian field,
Figure 3, were used in our simulation study. End-point linear
relative permeability curves were used for flow within
Relative permeability can be considered constant above the
fracture, see Figure 4. This end-point linear consideration is
bubble point, therefore, eqn (5) can be written as follows:
reasonable since linear type of relative permeability curves
pr 1 corresponds to a uniform distribution of pore sizes and
q = 1 J D kkro h dp (6) proppant in the fracture is a very uniform rock. Non-Darcy
p wf o Bo flow was considered negligible within this study. The
accuracy of the grid and the model was confirmed through
Correlations or approximations of the formation volume factor comparisons of single-phase results with analytical solutions.
and viscosity can be used to obtain the final form of the eqn
(6). However, taking the values of formation volume factor Construction of IPR Curves
and viscosity at mid-point between average reservoir pressure
and bottomhole pressure, eqn (4) can also be used. Figure 1 A simulation run is performed with the fractured well
shows the formation volume factor, viscosity, relative producing at constant rate until steady-state is reached. Once
permeability and mobility in the formation as function of the the steady-state is reached, the bottom flowing pressure is
reservoir pressure. It can be seen in this plot that the mobility recorded. This would give a point (q, pwf) of the IPR curve.
above the bubble point can be considered a straight line. This Many simulation runs are then conducted for a number of
suggests that the average mobility ratio above the bubble point rates with the same well conditions to obtain the IPR curve for
pressure can be evaluated at the mid-point between average the specific fracture conditions (CfD and Ix) and for the
reservoir pressure and bottomhole flowing pressure. specific PVT and relative permeability data. IPR curves are
generated for different combinations of fracture penetration
and fracture conductivity to evaluate the effect of CfD, and, Ix
SPE 94252 3

on the IPR curves. The same procedure is followed for the Another average correlation was also determined following
different set of rock and fluid properties, Siberians data and the same form as Vogels correlation, without considering CfD
Vogels. More simulations runs were performed with the data as a parameter. Eqn. (9) gives the final form of the correlation,
set from western-Siberia field than Vogels. However, both providing an error no greater than 5% for the same range of
provided very good representative amounts of runs to be use fracture conductivities.
in the correlations.
Due to the gas-bank effect, the beginning of steady-state time qo = qb +
J D pr
1.31
1 0.65 pwf 0.35
pb
( )
p wf 2
pb (9)
is different for each combination of fracture conductivity and
penetration. Therefore, for this study, the bottom flowing
pressure for all the runs were taken at the same areal Comparing Eqn (9) with Vogels, it can be seen that Vogel's
dimensionless time, tDA, that is, the time at which the fracture correlation underestimates fractured well performance below
with the highest conductivity reaches steady state (tDA=0.138). bubble point. Vogel suggests a correction of AOF by 45%
IPR curves for the fluid and rock properties described above meanwhile the simulation results (and Eqn (9)) show that the
are presented in Figure 5. These curves were normalized by correction should be only 22%. Therefore, engineer could
bubble point pressure and rate in order to compare the run have an error of 43% using Vogel for estimating AOF for a
results for different sets of data. Figure 6 presents the fractured well. Figure 6 shows the above remarks.
normalization of the IPR curves. As it can be seen in Figure 6, Additionally, eqn (9) is compared with the Vogel-Standing
only the data below the bubble point pressure is presented in correlation for two cases: a CfD of 0.5 and Ix of 0.1, and a CfD
the plot. of 2.0 and Ix of 0.3. For this comparison, the flow efficiency
was calculated from the CfD and Ix, and then IPR was
Simulation Results determined using Vogel-Standing correlation. Figures 9 and
10 show the results of this comparison. It can be seen that
Results of simulations for different fracture length and
Vogel-Standing correlation overestimates very much the IPR
conductivity at the same reservoir conditions are summarized
curves, to the point that makes the Standing correlation
in the Figure 5 and 6. It can be seen that the shape of the IPR
completely invalid for fractured wells.
dimensional curves depends on the fracture conductivity and
penetration. However, once the IPR curves are normalized, all Let us review now the production mechanisms that would
the IPR curves merge almost to a single curve. explain less multiphase effect in fractured wells compared to
radial-flow unfractured wells. It is known that fractured wells
Results of simulation of 5 PVT/Relative permeability data sets
have a different flow pattern due to the rock properties
have been combined in order to build a correlation, which is
differences between the fracture and the formation (difference
suitable for fractured wells in the range of conductivities
in conductivity). In order to understand more the multiphase
0.550 and fracture half-lengths xf = 100500 meters. The
flow effect in the fractured wells, two main factors were
correlation is presented is presented in eqn. (8).
analyzed. First, the mechanisms for the fact of having different
qo qb
qo , max qb
(
= 1 0.73 0.041Log C fd ( )) p wf
pr
(0.27 + 0.041Log (C fd ))(
p wf
pr
)2
relative permeability curves in the fracture than in the
formation. Secondly, the effect of reduction of the effective
qo , no skin qb permeability around the fracture due to the increase of gas
= 1.285 + 0.0347 Log (C fd ) saturation.
qo , max qb
The mobility behavior inside the fracture (using end-point-
(8) linear relative permeabilities) is presented in Figure 2. Note
the difference in fluid mobility below the bubble point
Eqn (8) suggests that the lower fracture conductivities is, the pressure compared to the mobility behavior in the formation as
straighter the curve is. The analysis also showed that the shown in Figure 1. In Figure 2, it can be seen that below the
multi-phase correction does not depend greatly on fracture bubble point pressure the mobility decreases almost linearly at
penetration but on fracture conductivity. The reason for this is slower pace than the mobility in the formation (comparing Fig
that the fracture conductivity is what dictates how big the gas 2 and Fig. 1). This implies that below bubble point, the
bank is created along the fracture. Figures 7 and 8 show the relative permeability or multiphase effect in the fracture is
gas bank generated for Ix =0.3 and CfD of 0.1 and 5, much less than the effect in the formation. Moreover, it can be
respectively. When Figure 7 is compared with Figure 8, it can seen that gas saturation in the fracture is very small compared
be seen that gas bank for a CfD of 0.1 is smaller than the gas to the one in the formation. Therefore, the fracture provides a
bank for CfD of 1.0. This can be explained with the pressure very conductive way for gas to escape from nearby formation
profiles generated for the two different cases of fracture and keeping relatively low gas saturation in the fracture and
conductivity. The lower conductivity case has higher pressure even in the formation (if it would be compared to radial
drop inside the fracture (for the same rate case) and therefore model).
the bubble pressure is reached closer to the wellbore than in
the higher-conductivity case. Therefore, lower conductivity Secondly, Holditch4 and many other authors have shown that
fractures would generate smaller bank effects. the production of a fractured well is not highly affected by
permeability reduction at the fracture face or fracture face
skin. One of the main reasons is that the fluid velocity across
4 SPE 94252

the fracture is a lot lower than in radial well model and the pb = Bubble point pressure, (bar)
pressure drop is directly proportional to the velocity.
However, it has also been shown that if the damage around the pr = Average reservoir pressure, (bar)
fracture is too deep, it could have significant production pwf = Wellbore flowing pressure, (bar)
impairment. In our case, the gas bank generates is big enough
to create some production impairment but not enough to be the qb = Oil flow rate at pwf = pb, (sm3/day)
same as in an unfractured well.
qo = Oil flow rate, (sm3/day)
Conclusions qo,max = absolute open flow, (sm3/day)
The following conclusions can be drawn from this work: qo,no skin = flow rate with no multiphase effect, (sm3/day)
1. Two new correlations to build IPRs curves or predict s = Skin factor (dimensionless)
production performance for fractured wells below the
bubble point pressure were presented in this paper. wf = Fracture width, (m)
2. Vogel's correlation underestimates fractured well xf = Fracture-half length (m)
performance below bubble point. Vogel suggests a
correction of AOF by 45% meanwhile the simulation o = Oil viscosity, (cp)
results and new correlation show that the correction
should be only 22%. Acknowledgements
3. Multiphase effect is dependent on fracture The authors thank YUKOS E&P management for allowing
conductivity and almost independent on fracture publishing this paper.
penetration. Higher conductivity fractures has bigger
gas banks therefore they are affected by multiphase References
effect to greater extent than lower conductivity
fractures. 1. Vogel J. V.: Inflow Performance Relationships for
Solution-Gas Drive Wells, Trans., AIME (1968)
4. Vogel-Standing correlation over-estimates the IPR 243, 83-92.
curve for fractured well if equivalent flow efficiency,
FE, is calculated from a fracture conductivity and 2. Standing, M.B. Inflow Performance Relationships
penetration. for Damaged Wells Producing by Solutions-Gas
Drive, Journal of Petroleum Technology (November
Nomenclature 1970) 1399-1400.
3. Craft B. C., Hawkins M.: Applied Reservor
t DA = Dimensionless time based on drainage area
Engineering, Englewood Clieffs, NJ: Prentice Hall,
rw = Wellbore radius (m) 1975.
4. Holditch, S.A.:Factors Affecting Water Blocking
re = Drainage radius (m)
and Gas Flow from Hydraulically Fractured Gas
Wells, Journal of Petroleum Technology (December
JD = Dimensionless Productivity Index
1979b) 31, No.12, 1515-1524.
Bo = Oil formation volume factor, (rm3/sm3)
Si Metric Conversion Factors
CfD = Dimensionless fracture conductivity
bbl * 1.589 873 E-01 = m3
FE = Flow efficiency, (dimensionless)
ft * 3.048 E-01 = m
hf = Fracture height, (m)
psi * 6.984 757 E+00 = kPa
Ix = Fracture penetration, (dimensionless)
k = Formation permeability, (mD)
kf = Proppant or fracture permeability, (mD)
ko = Effective permeability to oil, (mD)
krg = Relative permeability to gas, (dimensionless)
kro = Relative permeability to oil, (dimensionless)
Np = Proppant number, (dimensionless)
SPE 94252 5

1.8

1.6

1.4

1.2

0.8

0.6

0.4
0 25 50 75 100 125 150 175 200 225 250 275 300
Pressure (bar)

Mu_o, cp k_ro So Bo rm3/sm3 Normalised oil mobility

Figure 1 - Fluid mobility vs pressure (reservoir rock)

1.8

1.6

1.4

1.2

0.8

0.6

0.4
0 25 50 75 100 125 150 175 200 225 250 275 300
Block Pressure (bar)

Mu_o, cp k_ro So Bo rm3/sm3 Normalised oil mobility

Figure 2 - Fluid mobility VS pressure (within the fracture)

6 SPE 94252

2.5 70

60
2

50
Bo (rm3/sm3); o (cp)

1.5

Rs, sm3/sm3
40

30
1

20

0.5
10

0 0
0 50 100 150 200 250
Pressure, bar

Bo mu_o Rs

Figure 3 - PVT curves for the major western Siberia field used in analysis

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

krg (reservoir) kro (reservoir) krg (fracture) kro (fracture)

Figure 4 - Relative permeability curves for the major Western Siberia field
SPE 94252 7

300

250

200

150

100

50

0
0 50 100 150 200 250
Ix=0.1Fcd=5 Ix=0.1Fcd=10 Ix=0.1Fcd=20 Ix=0.1Fcd=50 Ix=0.1Fcd=70 Ix=0.1Fcd=100 Ix=0.1Fcd=200 Ix=0.1Fcd=500 Ix=0.1Fcd=1000 Ix=0.2Fcd=0.1
Ix=0.2Fcd=0.5 Ix=0.2Fcd=1 Ix=0.2Fcd=2 Ix=0.2Fcd=5 Ix=0.2Fcd=10 Ix=0.2Fcd=20 Ix=0.2Fcd=50 Ix=0.2Fcd=70 Ix=0.2Fcd=100 Ix=0.2Fcd=200
Ix=0.2Fcd=500 Ix=0.2Fcd=1000 Ix=0.3Fcd=0.1 Ix=0.3Fcd=0.5 Ix=0.3Fcd=1 Ix=0.3Fcd=2 Ix=0.3Fcd=5 Ix=0.3Fcd=10 Ix=0.3Fcd=20 Ix=0.3Fcd=50
Ix=0.3Fcd=70 Ix=0.3Fcd=100 Ix=0.3Fcd=200 Ix=0.3Fcd=500 Ix=0.3Fcd=1000 Ix=0.6Fcd=0.1 Ix=0.6Fcd=0.5 Ix=0.6Fcd=1 Ix=0.6Fcd=2 Ix=0.6Fcd=5
Ix=0.6Fcd=10 Ix=0.6Fcd=20 Ix=0.6Fcd=50 Ix=0.6Fcd=70 Ix=0.6Fcd=100 Ix=0.6Fcd=200 Ix=0.6Fcd=500 Ix=0.6Fcd=1000 Ix=1Fcd=0.1 Ix=1Fcd=0.5

Figure 5 IPR curves results from the simulation

0.9

0.8

0.7

0.6

0.5

qo qb
=
1 0.65
p wf
pr 0.35 ( )
p wf 2
pr
0.4
qo , no skin qb 1.31
Rueda Zakharov

0.3 qo qb
=
1 0.2
pw f
pr 0.8 ( )
p wf 2
pr

qo , no skin qb 1.8
Vogel
0.2

0.1

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Ix=0.1Fcd=5 Ix=0.1Fcd=10 Ix=0.1Fcd=20 Ix=0.1Fcd=50 Ix=0.1Fcd=70 Ix=0.1Fcd=100

Ix=0.1Fcd=200 Ix=0.1Fcd=500 Ix=0.1Fcd=1000 Ix=0.2Fcd=0.1 Ix=0.2Fcd=0.5 Ix=0.2Fcd=1
Ix=0.2Fcd=2 Ix=0.2Fcd=5 Ix=0.2Fcd=10 Ix=0.2Fcd=20 Ix=0.2Fcd=50 Ix=0.2Fcd=70
Ix=0.2Fcd=100 Ix=0.2Fcd=200 Ix=0.2Fcd=500 Ix=0.2Fcd=1000 Ix=0.3Fcd=0.5 Ix=0.3Fcd=1
Ix=0.3Fcd=2 Ix=0.3Fcd=5 Ix=0.3Fcd=10 Ix=0.3Fcd=20 Ix=0.3Fcd=50 Ix=0.3Fcd=70
Ix=0.3Fcd=100 Ix=0.3Fcd=200 Ix=0.3Fcd=500 Ix=0.3Fcd=1000 Vogel Rueda-Zakharov

Figure 6 - Simulation results normalized by pb and qo, no skin (only points below pb are shown)
8 SPE 94252

Figure 7 - Gas bank generated for a fractured well of IX = 0.3 and CfD = 0.1

Figure 8 - Gas bank generated for a fractured well of IX = 0.3 and CfD = 1.0
SPE 94252 9

0.9

0.8

0.7

0.6
p w /p r

0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1

q o /q Darcy

Standing Rueda-Zakharov Standing approx Vogel

Figure 9 Comparison of Vogel-Standing vs Rueda-Zakharovs correlations for Ix=0.1 and Cfd=0.5

0.9

0.8

0.7

0.6
p w /p r

0.5

0.4

0.3

0.2

0.1

0
0 0.2 0.4 0.6 0.8 1

q o /q Darcy

Standing Rueda-Zakharov Standing approx Vogel

Figure 10 Comparison of Vogel-Standing vs Rueda-Zakharovs correlations for Ix=0.1 and Cfd=2.0