SPE-172928-MS Analysis of Transient Linear Flow Associated With Hydraulically-Fractured Tight Oil Wells Exhibiting Multi-Phase Flow

Download as pdf or txt
Download as pdf or txt
You are on page 1of 18

SPE-172928-MS

Analysis of Transient Linear Flow Associated with Hydraulically-Fractured


Tight Oil Wells Exhibiting Multi-Phase Flow
H. Behmanesh,H. Hamdi, and C. R. Clarkson, University of Calgary

Copyright 2015, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Middle East Unconventional Resources Conference and Exhibition held in Muscat, Oman, 26 –28 January 2015.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper 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 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
Hydraulically-fractured vertical and horizontal wells completed in the tight formations typically exhibit
long periods of transient linear flow that may last many years or decades. From this transient linear flow
period, the linear flow parameter (xf公k) may be extracted. However, changes in effective permeability
to the oil phase during production, caused by wellbore pressure falling below the saturation pressure,
affect the flow dynamics in tight oil reservoirs and complicate the analysis. The use of methods that
assume single-phase flow properties, such as the square-root of time plot, can lead to significant errors in
linear flow parameter estimates.
In this study, an analytical method is introduced to mathematically correct the slope of the square-
root-of-time plot for the effects of multi-phase flow through the use of modified pseudovariables.
Although the correction was derived for wells producing at constant flowing pressure during transient
linear flow, the method is extended for wells producing at variable rate/flowing pressures. In order to
evaluate pseudovariables used in the correction, the saturation-pressure relationship must be known. In
this work, an analytical method for evaluating the saturation-pressure relationship is also developed.
The results of our new analytical method for linear flow analysis are validated against numerical
simulation. The new method yields linear flow parameter estimates that are within 10% of those input into
the numerical simulator.
Introduction
Commercialization of low-permeability (tight) reservoirs producing oil is now possible due to new
technology, such as multi-fractured horizontal wells (MFHW). It is desirable to be able to apply
rate-transient analysis methods to producing tight oil wells for the purpose of extracting hydraulic fracture
and reservoir properties, however current methods typically assume single-phase flow in the reservoir.
Further, existing methods were developed for conventional reservoirs; there are substantial differences in
reservoir performance characteristics between conventional and ultra-low permeability reservoirs that can
render existing methods inapplicable.
Transient linear flow is a very important flow-regime in tight formations and is associated with flow
of reservoir fluids toward hydraulically-fractured wells. Different investigators have studied two-phase
flow problems over the last few decades (Raghavan, 1976, Camacho and Raghavan, 1989, Bøe et al.,
2 SPE-172928-MS

1989), but the focus has been primarily on conventional reservoirs; nonlinearities associated with
two-phase flow in tight formations have not been dealt with extensively. Traditionally, for conventional
reservoirs, pressure- and saturation-dependent properties have been accounted for by using a two-phase
pseudopressure. However, more rigorous treatment is required to account for high saturation gradients
near the wellbore experienced in tight reservoirs. Two-phase pseudotime is also required to account for
pressure-dependent properties of the fluids, and to address flow-equation nonlinearities.
In this work, we have developed a rate transient analysis (RTA) technique for analyzing tight oil
reservoirs during the transient linear flow period. In previous work (Behmanesh et al. 2015), we used an
analogous approach for analyzing tight gas condensate reservoirs, however there are several important
differences encountered for the case of black oil. Firstly, it is the oil component which is the focus of the
current study, and efforts to linearize the diffusivity equation focus on the oil phase. The substantial
increase in oil compressibility during the transition from single-phase to two-phase flow adds additional
nonlinearities not dealt with in the gas condensate problem. Secondly, unlike the gas condensate case
where condensate remains relatively immobile for shale reservoirs, in the black oil case, the high oil
saturation ensure some mobility of that phase in the reservoir. As with the gas condensate case, however,
our method for analysis includes the evaluation of two-phase pseudopressure and two-phase pseudotime
which are used to linearize the diffusivity equation. The resulting solution is then applied to simulated
cases for the purpose of extracting the linear flow parameter.
The significance of this work is that we have proven that the liquid solution analog can be applied
(through the use of carefully defined pseudovariables) for low-permeability oil reservoirs. Several
commonly observed behaviors of tight oil wells are also explained analytically using our derived
equations.
The manuscript is organized as follows: first, we discuss the effect of well geometry (radial or linear
model), reservoir properties (reservoir permeability) and producing boundary conditions (constant rate or
constant flowing bottomhole pressure) on the transient behavior of oil reservoirs, which is mainly
reflected in the GOR behavior. Then, the new mathematical model for analysis of production data is
presented. In the “Saturation-Pressure Relationship” section we discuss different methods for estimating
the saturation-pressure relationship that is required for evaluation of pseudovariables. Application of our
new methods for establishing the linear flow parameter (xf公k) is demonstrated in the “Example
Applications” section.
Gas-Oil Ratio Behavior Under Various Production Conditions
Aanonsen (1985) applied the similarity method in the study of multiphase flow for solution-gas-drive and
gas condensate reservoirs. He applied the method to constant rate radial flow problems for gas condensate
and solution-gas-drive reservoirs, and demonstrated that pressures and saturations are unique functions of
the Boltzmann variable (O’Sullivan, 1981). These observations have also been documented by other
investigators (Bøe, 1989; Serra, 1988). Bøe et al. (1989) concluded that the producing gas-oil ratio (GOR)
quickly stabilizes and remains constant throughout the infinite acting period. A reservoir permeability of
10 md was used in simulation studies of Bøe et al. (1989). A constant GOR, as well as a unique
saturation-pressure relationship, has also been reported for the case of constant flowing pressure transient
linear flow for two-phase tight oil and gas condensate cases (Whitson and Sunjerga, 2012, Behmanesh et
al., 2015 and Tabatabaie, 2014). Accordingly, in the following we have performed a systematic study of
GOR behavior under various production conditions, which will assist in understanding reservoir perfor-
mance and provide guidance for production data analyses of multi-phase flow cases.
In this section, the simulated production performance of a volatile oil reservoir with an initial solution
gas oil ratio (Ri) of 1090 scf/STB is studied. Fluid property variation with pressure is shown in Figure 1.
Our objective is to investigate the effect of different reservoir geometries and reservoir permeabilities (100
nd to 100 md) on the producing GOR for wells subject to constant oil rate and constant flowing
SPE-172928-MS 3

Figure 1—Fluid properties of the volatile oil used in numerical simulation.

bottomhole pressure (FBHP) conditions. The initial pressure for all the simulated cases is equal to the
bubble point pressure (4032 psia).
Effect of Inner Boundary Condition. The dimensionless pressure profile for a single-phase liquid
during infinite-acting linear flow for constant FBHP production is given in Eq. 1 (Wattenbarger et al.
1998). Eq. 2 gives the pressure profile for the infinite-acting radial flow case subject to constant surface
rate production:
(1)

(2)

For the constant rate and constant FBHP condition at the wellbore, the dimensionless pressures are
defined as pD⫽kh(pi-p)/141.2qB␮ and pD⫽(pi-p)/(pi-pwf), respectively.
The pressure profiles for both of the solutions (constant FBHP linear and constant rate radial) are only
a function of Boltzmann variable, xD/公td. Thus, by assuming that the Boltzmann transformation is valid
for two-phase flow for these cases, a unique pressure-saturation (P-S) relationship can be found in the
absence of a mechanical skin factor. This finding has been verified with numerous compositional and
black oil simulations. Another important observation is that the constant producing GOR during infinite-
acting flow may last many years or decades. For shale gas condensate systems, the cause of constant GOR
during transient linear flow (and constant FBHP) is thought to be due to the fact that practically all liquid
dropout in the reservoir will remain unproduced (essentially zero mobility), and the producing GOR is
therefore mostly dependent on the FBHP (Whitson and Sunjerga, 2012). For shale oil reservoirs, however,
oil will always be mobile in the reservoir due to higher initial oil phase saturations.
Bøe et al. (1989) presented an analytical solution to predict constant GOR behavior for the case of the
constant rate boundary condition. Later, Behmanesh et al. (2015) provided a similar solution predicting
constant producing GOR for tight oil and gas condensate cases subject to constant FBHP for the 1D planar
geometry. In order to investigate whether these observed GOR behaviors are due to inner boundary
conditions (i.e. constant pressure for planar slab fracture and constant rate for radial vertical well) or
reservoir properties, a sensitivity to permeability was performed. The resulting GOR profile for different
reservoir permeability values for the constant rate, radial flow case is given in Figure 2. The surface oil
producing rate for k⫽100 md is 8000 STB/D. For the lower permeability cases, the rate is reduced to a
value so that the FBHP for all the simulated cases are comparable. The production rate for the case with
k⫽0.01 md is 1.95 STB/D. The producing GOR stabilizes at the same level, independent of reservoir
4 SPE-172928-MS

Figure 2—Producing GOR versus time for constant surface oil rate, radial flow case for different reservoir permeability values.

Figure 3—The producing GOR for constant FBHP (1000 psia), 1D planar fracture geometry case for different reservoir permeability values.

permeability, as long as the FBHP drops with the same rate. A comparable plot was prepared for constant
FBHP, planar slab fracture case (Figure 3). For this case, GOR stabilization is also independent of
permeability level.
Effect of Flow Geometry. In liquid rich shale (LRS) reservoirs, the extra low permeability of the
porous medium imposes a high pressure drop, which leads to large gas-to-oil mobility ratio gradients near
the fracture face. This results in leaner wellstreams for LRS cases compared to conventional reservoirs for
the same fluid system (Whitson and Sunjerga, 2012). For conventional reservoirs (i.e. radial geometry),
the dimensionless producing oil-gas ratio (rpD⫽Ri/rp) has typical values around unity during the infinite-
acting flow period. However, infinite-acting flow behavior of LRS wells is usually characterized by rpD
ranging from 0.05-0.5 for volatile oils (Ri⬎1000 scf/STB) and 0.5-0.9 for lower-GOR/undersaturated oils
(Whitson and Sunjerga, 2012).
To study the behavior of rpD for different reservoir/well geometries, black oil numerical modeling
results of a vertical well (radial model) and a horizontally fractured well (linear model) are compared. A
reservoir permeability of 0.1 md is used in both simulation models. GOR profiles for both of the
geometries are given in Figure 4. As expected, the rpD for the radial model is close to 1 (meaning that the
producing wellstream is almost identical to the initial reservoir fluid) and for the linear model it is close
SPE-172928-MS 5

Figure 4 —Simulated OGR profile for a tight oil reservoir producing at constant FBHP (1000 psia) using a 1D planar fracture and radial geometry.
For this simulation run, Soiⴝ1; and relative permeability curves with corey exponents of 2.5 for both oil and gas phase is used (piⴝpbⴝ4033 psia,
Riⴝ1090 scf/STB).

Figure 5—Simulated oil saturation profile corresponding to Figure 4.

Figure 6 —The producing GOR for constant FBHP (1000 psia), radial flow case for different reservoir permeability values. The producing fluid is
similar to initial fluid because there exists a logarithmic pressure profile around the wellbore and a greater fluid volume passes through the near well
region than for the linear flow case.
6 SPE-172928-MS

to 0.25, which is consistent with the ranges provided


by Whitson and Sunjerga (2012). Similar results
have been observed for other reservoir permeability
values (ranging from 100 nd to 100 md).
Saturation profiles corresponding to Figure 4 are
shown in Figure 5. Oil saturation for the radial
model is higher than for the linear model (with all
other fluid and rock properties being the same ex-
cept). This is mainly because, for the radial model,
a logarithmic pressure profile exists around the Figure 7—The average pressure in the distance of investigation is con-
wellbore and a greater fluid volume passes through stant for a well with a planar slab fracture producing with constant
FBHP.
the near well region resulting in higher oil mobility.
This statement has been validated by many numerical simulations for different permeability values in both
radial and linear models. The GOR versus time plot for the radial flow case is given in Figure 6 (compare
with the linear flow case in Figure 3).
Assuming that the Boltzmann transformation is valid during infinite-acting radial flow for a constant
production rate (and also during infinite-acting linear flow with constant FBHP), it can be mathematically
shown that the (volumetric) average of any property is constant in the distance of investigation. This is
irrespective of reservoir properties and is the manifestation of the nature of the governing partial
differential equation and corresponding boundary conditions. This is illustrated schematically for pressure
in Figure 7. Proof of this observation is given mathematically in Appendix A.

Mathematical Model for Production Analysis of Tight Oil Reservoirs


The governing PDEs (Formation Volume Factor formulation) for oil and gas components in one-
dimensional Cartesian coordinates are written as follows:
(3)

(4)

Capillary and gravity effects are ignored in Eq. 3 and Eq. 4. In order to write the equations in a more
compact fashion, the following notations are used: S⫽So, a⫽k ខ rg/␮gBgd⫹Rskro/␮oBo, ␣⫽ Rvkrg/
␮gBgd⫹Rskro/␮oBo, b⫽ Sg/Bgd⫹So/Bo, ␤⫽ RvSg/Bgd⫹So/Bo and x ⬅ (⭸x/⭸S)p and x’ ⬅ (⭸x/⭸p)S and x僆
(a, ␣, ␤, b).Eq. 3 may then be written as:
(5)

The nonlinearities in Eq. 5 represented by coefficient ‘␣’ can be eliminated by introducing an integral
pseudopressure transformation as follows:
(6)

To calculate the pseudopressure, a relationship between saturation and pressure is required in order to
integrate relative permeability over pressure. This is discussed further in the “Saturation-Pressure
Relationship” section. Use of pseudopressure alone, however, cannot fully linearize the diffusivity
equation. An empirical pseudotime transformation is also required to linearize the temporal portion of the
partial differential and is given as follows:
(7)
SPE-172928-MS 7

Therefore a linearized form of the diffusivity equation for two-phase flow conditions, using 2-phase
pseudopressure and pseudotime, can be written as follows:
(8)

For boundary-dominated flow, the integrand parameters in Eq. 7 are evaluated at an average reservoir
pressure (Heidari and Gerami, 2011). During the boundary-dominated flow regime, average reservoir
pressure declines with production time. For transient flow, however, we use the average pressure in the
distance of investigation (DOI) as proposed by Anderson and Mattar (2007). Nobakht and Clarkson
(2012) and Behmanesh et al. (2014) utilized this approach for dry gas reservoirs and gas condensate
reservoirs, respectively. The average pressure in the DOI tends to be a constant value for constant flowing
pressure as demonstrated in the Appendix A. Conversely, for constant rate production during transient
linear flow, both average pressure and oil saturation are reduced in the growing DOI over time.
Constant-Pressure Solution. For constant flowing pressure production, the solution to Eq. 8 is given
by Wattenbarger et al.(1998) as follows:
(9)

Eq. 9 can be simplified to the form of 1/qo⫽m’cp公ta,tp where “m’cp” is the slope of characteristic plot
of 1/qo versus 公ta,tp. Further, xf公k can be calculated from the following equation:
(10)

Calculation of pseudotime in Eq. 7 can be simplified by noting that average pressure in the DOI for
the constant flowing pressure case is constant:
(11)

Therefore Eq. 10 can be written as:


(12)

where, fcp is a correction factor and is given as follows:


(13)

Constant-Rate Solution. For constant rate production, the solution to Eq. 8 takes the following form
(Wattenbarger et al., 1998):
(14)

Eq. 14 can be simplified to the form of (Ppi-Ppwf)/qo⫽m’cR公ta,tp, where “m’cR” is the slope of
characteristic plot of (Ppi-Ppwf)/qo versus 公ta,tp. Further, xf公k can be calculated from the following
equation:
(15)

Variable Rate/Variable Pressure Conditions. Historically, superposition time functions (bilinear,


linear, radial, boundary-dominated flow) are used to account for changing rate with time. By applying
superposition principles, the variable rate data can be effectively converted to an equivalent constant rate
condition. However, because of pressure-dependent PVT properties, (corrected) pseudotime (tatp,SL)
should be used instead of actual time. The corrected two-phase linear superposition pseudotime function
is calculated as:
8 SPE-172928-MS

(16)

To account for the stress-sensitivity of reservoir permeability, pseudopressure and pseudotime are
modified accordingly and given as:
(17)

(18)

Saturation-Pressure Relationship
A saturation-pressure relationship is required to relate relative permeability to pressure in order to evaluate
the corresponding pseudopressure integral. The saturation-pressure relationship may or may not change
during production life, depending on the particular flow period (transient flow period or depletion flow
period), the flowing boundary conditions (constant pressure production or constant rate production) and
flow geometry (radial model or linear model). In this section, some methods for estimating the P-S path
are presented.
Unsteady-state path. The unsteady-state path is related to the continuously changing overall compo-
sition entering and exiting the control volume in absence of mechanical skin. Bøe et al. (1989) developed
a method to compute the saturation-pressure path analytically for two-phase radial flow for a well subject
to constant surface rate production. This path is used in the two-phase drawdown model. The equation was
developed by combining the partial differential equations for oil and gas components through the use of
the Boltzmann variable. Behmanesh et al. (2013) developed a similar equation for two-phase linear flow
conditions for wells producing at constant FBHP:
(19)

For constant FBHP, the auxiliary rate equation applicable to the constant rate condition is not useful.
The limiting cases of Eq. 17, however, provide very valuable information to approximate the full path of
pressure and saturation. The two limiting cases which represent long production time and early production
time are: Limiting Case I: ds/dp ⫽ (␣a= ⫺ a␣=)/(a␣˙ ⫺ ␣ȧ) and Limiting Case II: ds/dp ⫽ (␣b= ⫺
a␤=)/(a␤˙ ⫺ ␣ḃ) respectively. The combination of the two limiting cases and a cubic polynomial can be
used to approximate the saturation-pressure path - the details for obtaining the polynomial coefficients
have been documented in Behmanesh et al. (2015). The use of a cubic polynomial for P-S path prediction
yields negligible error in the calculation of pseudopressure. Figure 8 provides plots for different
saturation-pressure paths for one fluid model and relative permeability curve used in this work. The
saturation-pressure relationship for constant pressure production is calculated from unsteady-state path.
Analysis of Eq. 19 yields the following important results. We see that the producing GOR is given by
R⫽a/␣. The total derivative of R with respect to pressure is:
(20)

Substituting the Limiting Case I expression for ds/dp ds/dp ⫽ (␣a= ⫺ a␣=)/(a␣˙ ⫺ ␣ȧ), into this
derivative we find dR/dp⫽0; i.e., the GOR is a constant. This proves analytically the observation of a
constant GOR during linear transient period for all LRS producing against constant pressure, which was
previously found through numerical simulation studies. This further implies that a steady state region,
although very small, is developed near the wellbore.
Steady State Path. The steady-state path describes the saturation-pressure route for a fluid of fixed
composition. This method is computationally very attractive and simple. Evinger and Muskat (1942) were
among the first to propose this model. The producing GOR is used to find the relationship between the
SPE-172928-MS 9

Figure 8 —Saturation-pressure relationship. The numerical simulation-derived curve, interpolation using polynomial function, steady state, and
limiting case II are also shown.

Figure 9 —Saturation-pressure relationship for a well producing at constant surface rate production after 1 year and 2 years of production during
transient linear flow period.

saturation and pressure at the sandface which is assumed to be applicable for the whole reservoir
(Raghavan, 1976, Fetkovich et al. 1986). The GOR equation is given as:
(21)

The saturation-pressure path can be calculated through re-arrangement of GOR equation as follows:
(22)

The GOR equation is equivalent to the Limiting Case I discussed earlier. Use of the steady state
assumption to predict the full P-S path for tight formations is however incorrect. As shown in the Figure
8, for the constant pressure case, the GOR equation cannot be applied to predict the saturation-pressure
relationship for pressures up to bubble point pressure because the steady state region is not extensive in
tight formations. Equation 18 is applicable when both gas and oil are flowing in the reservoir i.e. the gas
saturation surpasses the critical value. However, a combination of Equation 18 and Constant Composition
Experiment (CCE) can be used to evaluate the full saturation-pressure relationship (Qanbari and Clarkson,
2013). Numerical simulations confirming this result are given in Figure 9. The GOR relationship in
10 SPE-172928-MS

combination with the CCE test is used to calculate Table 1—Rock and fluid properties for Example 1.

saturation-pressure relationship for constant rate Properties Value


and variable rate/pressure boundary conditions. Rock compressibility 6.8 E-06 psi⫺1
CCE path. The information obtained from lab- Permeability modulus 8.0 E-05 psi⫺1
oratory experiments may be used predict the satu- Reservoir permeability 0.01 md
ration-pressure relationship. The CCE experiment Reservoir porosity 0.10
Initial reservoir pressure 4033 psia
provides information about the relative amount of Bubble point pressure 4033 psia
oil and gas. The test simulates primary depletion of FBHP 1000 psia
a specific fluid at a constant temperature. The over- Boi 1.2 RB/STB
all composition remains constant during a CCE test Coi 2E-05 1/psia
Reservoir thickness 100 ft
and can be used to predict the saturation-pressure
xf 250 ft
path for pressure regions which are just below sat- Initial water Saturation 0
uration pressure (where gas has evolved out of Corey exponent no⫽ng 2.5
solution, but where its saturation is less than critical
saturation to flow). The saturation can be correlated
to PVT properties as given in Eq. 23.
(23)

The derivation of Eq. 20 is given in Appendix B.


The CCE test results are very close to the result of
limiting case II from the unsteady state path.
Tank model. The saturation-pressure path given
by the tank model is actually the relationship be-
tween average pressure and average saturation in
the reservoir and is given from a material balance
(Eq. 24). This equation is used to evaluate the
average saturation in the distance of investigation
Figure 10 —Square-root-of-time plot for Example 1. This figure illus-
for calculation of pseudotime (for constant rate and trates the effect of corrected pseudotime for multiphase flow on the slope
variable rate/pressure boundary conditions). of square-root-of-time plot.

(24)

Examples
In order to check the validity of the approaches described in this paper, a series of fine-grid black oil
simulations were conducted. Eclipse 100 (Schlumberger, 2010) was used for the reservoir simulator. The
simulations were also used to study the performance of typical tight oil reservoirs discussed in previous
sections. The details of simulation model set up and gridding is discussed elsewhere (Behmanesh et al.,
2015). In this section, three synthetic examples of constant pressure, constant rate and variable pressure/
rate production data during the transient linear flow period are analyzed.
Constant-Pressure. For this case, a 1D model for a hydraulically-fractured well is used to simulate
production from a saturated oil reservoir. The initial reservoir pressure is set equal to the fluid saturation
pressure of 4033 psia. Other pertinent rock and fluid properties are given in Table 1. The procedure for
analyzing transient linear flow for the constant pressure case is summarized below.
1. Plot inverse oil rate versus square-root-of-time in Cartesian coordinates and determine the slope
of line, mCP. The data should follow a straight line trend during linear flow.
2. Determine xf公k using the initial properties, with Eq. 12 (fCP⫽1).
SPE-172928-MS 11

Figure 11—Square-root-of-time plot for Example 2.

3. Evaluate the average pressure in the region of influence ( ) from a material balance equation. The
material balance equation is provided in Appendix C. The average pressure in the region of
influence is constant.
4. Evaluate the correction factor, fCP, with Eq. 13 (now accounting for multi-phase flow), and
multiply the xf公k from step 2 by fCP.
5. Compare the xf公k from step 4 and step 2. If the changes are sufficiently small, the convergence
criterion is satisfied. Otherwise, return back to step 3 and follow step 4 and 5.
A plot of 1/qo versus 公t and 公ta,tp for this case is shown in Figure 10. The calculated xf公k from Eq.
12 is 26.4 ft.md-0.5, which is within 5.4% of the value input into the numerical simulator. The calculated
fcp (Eq. 13) is 0.57. The uncorrected xf公k is 37, which means that an 68% error will result if multiphase
flow and stress sensitive permeability effects are not accounted for.
Constant-Rate. This example is similar to the previous example, except that the well constraint is
constant-rate production, the permeability modulus is assumed to be zero, and k⫽0.05 md. The well is
producing at 24 STBD for three years. For constant rate production, the initial estimate of xf公k is given
as:
(25)

where mcR is the slope of (Ppi-Ppwf)/qo versus 公t. The procedure for analyzing transient linear flow for
the constant rate is summarized as follows:
1. Plot Ppi-Ppwf versus 公t in Cartesian coordinates and determine the slope of early-time data points
(mCR) and calculate xf公k using Eq. 22.
2. Evaluate average pressure and average saturation in the region of influence from the material
balance equation. The average pressure and saturation in the distance of investigation decrease
over time.
3. Determine the corrected pseudotime for constant rate production from Eq. 7.
4. Plot Ppi-Ppwf versus 公tatp in Cartesian coordinates and find the slope of early-time data and
recalculate xf公k according to Eq. 15.
5. Repeat steps 2-4 until xf公k converges.
A plot of 1/qo versus 公tatp for this case is shown in Figure 11. The calculated xf公k from Eq. 15 is
54 ft.md⫺0.5 which is within 5% of the value input into the numerical simulator, 55.9 ft.md⫺0.5. The
uncorrected xf公k is 75, ft.md⫺0.5 which means that an 34% error will result if multiphase flow effects
are not accounted for.
12 SPE-172928-MS

Figure 12—Average pressure and average saturation in the DOI for Example 2.

Figure 13—FBHP and production rate for Example 3.

The average pressure and the average saturation in the DOI for Example 2 is plotted in Figure 12.
Variable Rate-Variable Pressure. This example is similar to Example 2, except that the well is
subject to a variable flowing pressure constraint (and hence has variable production rate). A plot of FBHP
and oil rate is given in Figure 13. The analysis procedure is similar to the constant rate case, except that
the analysis is carried out with linear superposition pseudotime, with two-phase flow accounted for (Eq.
16). The calculated xf公k is 61 ft.md⫺0.5 which is within 9% of the value input into the numerical
simulator, 55.9 ft.md⫺0.5. The uncorrected xf公k is 77, which means that an 37% error will result if
multiphase flow effects are not accounted for.
Summary and Conclusions
In this study, the effect of well geometry (radial versus linear flow), permeability and inner well boundary
conditions on GOR characteristics of tight oil reservoirs is discussed. One particularly important
observation is that, for the range of reservoir permeabilities investigated (from 100 nd to 100 md), the
producing GOR stabilized at the same level in all cases, regardless of reservoir permeability for both radial
and linear flow. For the radial model, the dimensionless producing oil-gas ratio is around unity, but is
lower for the linear flow model. This is mainly because, for the radial model, a logarithmic pressure
profile exists around the wellbore, resulting in sufficient fluid volume passing through the near well
region, leading in turn to higher oil mobility.
SPE-172928-MS 13

A new mathematical model for analysis of production data from tight oil reservoirs is also presented
for different operating conditions. For constant flowing bottomhole pressure production, a correction
factor, fCP, is given. The primary goal of this correction factor is to correct the slope of the square-root
of time plot for the effects of multi-phase flow. Modified pseudovariables are required to make this
correction – these pseudovariables result in a linearized diffusivity equation, allowing the solution for
liquids to be used. These corrections result in a much-improved estimate of xf公k in the presence of
multi-phase flow. One complication is that, for pseudovariable calculation, the saturation-pressure
relationship must be provided - different methods for estimating the P-S path are therefore presented. It
is further proved analytically that, whenever the Boltzmann transformation is valid (constant surface rate
for radial models and constant FBHP for linear models), average properties (e.g. saturation and pressure)
in the distance of investigation are constant.

Acknowledgments
The authors would like to thank the sponsors of Tight Oil Consortium for funding this project. Chris
Clarkson would like to acknowledge Shell, Encana and Alberta Innovates Technology Futures for support
of his Chair in Unconventional Gas and Light Oil Research in the Department of Geoscience, University
of Calgary. Hamidreza Hamdi thanks the Interactive Reservoir Modeling, Visualization and Analytics
Research Group at the University of Calgary for supporting his postdoctoral fellowship. Finally, the
authors would like to thank Dr. Curtis H. Whitson from NTNU University for fruitful discussions on the
subject of rate-transient analysis.

Nomenclature
Bgd ⫽ “dry” gas formation volume factor, ft3/scf or RB/Mscf.
Bo ⫽ oil formation volume factor, RB/STB.
CCE constant composition experiment
DOI ⫽ distance of investigation
FBHP ⫽ flowing bottomhole pressure
FCD ⫽ dimensionless fracture conductivity.
GOR ⫽ gas oil ratio, Mscf/STB.
k ⫽ formation permeability, md or nd.
kf ⫽ fracture permeability, md.
krg ⫽ gas relative permeability.
kro ⫽ oil relative permeability.
Lh ⫽ length of horizontal well, ft.
LRS ⫽ Liquid Rich Shale.
n ⫽ grid block index.
OGR ⫽ oil gas ratio, STB/MMscf.
p
⫽ pressure, psia.
pb ⫽ based pressure, psia
pi ⫽ initial pressure, psia.
⫽ average pressure, psia
pwf ⫽ flowing bottomhole pressure (FBHP), psia.
qg ⫽ Surface gas rate, scf/D, Mscf/D, MMscf/D.
qo ⫽ Surface oil rate, STB/D.
rp ⫽ producing oil-gas ratio or liquid yield, STB/MMscf.
Rs ⫽ solution gas-oil ratio (GOR), scf/STB, Mscf/STB.
Rsi ⫽ initial solution gas-oil ratio (GOR), scf/STB, Mscf/STB
14 SPE-172928-MS

RTA ⫽ rate transient analysis


t ⫽ time, hrs, days.
xf ⫽ Fracture half-length, ft.
␮g ⫽ gas viscosity, cp.
␮o ⫽ oil viscosity, cp.

References
Anderson, D. and Mattar L. 2007. An improved pseudo-time for gas reservoirs with significant
transient flow. Journal of Canadian Petroleum Technology 46.(7). SPE-PETSOC-07-07-05.
http://dx.doi.org/10.2118/07-07-05.
Behamensh, H., Tabatabaie, S. H., Heidari, M and Clarkson, C. R. 2014. Modification of the Transient
Linear Flow Distance of Investigation Calculation for Use in Hydraulic Fracture Property Deter-
mination. Paper SPE 168981 presented at the SPE Unconventional Resources Conference, The
Woodlands, Texas, USA, 1-3 April. http://dx.doi.org/10.2118/168981-MS.
Behmanesh, H., Hamdi, H., Clarkson, C. R. 2015. Production data analysis of tight gas condensate
reservoirs, 22, 22–34. Journal of Natural Gas Science and Engineering, http://dx.doi.org/10.1016/
j.jngse.2014.11.005.
Bøe, A., Skjaeveland S., and Whitson C. 1989. Two-phase pressure test analysis. SPE Form Eval 4(4):
604 –610. SPE-10224-PA. http://dx.doi.org/10.2118/10224-PA.
Camacho-v, R. G., and Raghavan R. 1989. Performance of wells in solution-gas-drive reservoirs. SPE
Form Eval 4.(4) 611–620. SPE-16745-PA. http://dx.doi.org/10.2118/16745-PA.
Dake, L. P. (1978). Fundamentals of reservoir engineering. Elsevier.
Evinger, H.H. and Muskat, M. 1942. Calculation of Theoretically Productivity Factor. Trans. 146(1):
126 –139. http://dx.doi.org/10.2118/942126-G.
Fetkovich, M. J. (1973). The Isochronal Testing of Oil Wells. SPE paper 4529 presented at the 48th
Annual Fall Meeting of the Society of Petroleum Engineers of AIME, Las Vegas, Nev., 30
September – 3 October.
Heidari, M., Gerami, S. 2011. A New Model for Modern Production-Decline Analysis of Gas/
Condensate Reservoirs, Journal of Canadian Petroleum Technology 50(7-8): 14 –23. SPE-
149709-PA. http://dx.doi.org/10.2118/149709-PA.
Nobakht, M. and Clarkson, C. 2012a. A New Analytical Method for Analyzing Linear Flow in
Tight/Shale Gas Reservoirs: Constant-Flowing-Pressure Boundary Condition. SPE Res Eval &
Eng 15(3), 370 –384. SPE-143989-PA. http://dx.doi.org/10.2118/143989-PA.
Qanbari, F. and Clarkson, C.R. 2013. Analysis of Transient Linear Flow in Tight Oil and Gas
Reservoirs with Stress-Sensitive Permeability and Multi-Phase Flow Paper SPE 167176 presented
at the SPE Unconventional Resources Conference, The Woodlands, Texas, USA, 1-3 April.
http://dx.doi.org/10.2118/168980-MS.
Raghavan, R. 1976. Well Test Analysis: Wells Producing by Solution Gas Drive. SPE J. 16(4): pp.
196 –208. SPE-5588-PA. http://dx.doi.org/10.2118/5588-PA.
O’Sullivan, M. J. (1981). A similarity method for geothermal well test analysis. Water Resources
Research, 17(2), 390 –398.
Schlumberger, 2010, Eclipse 100 and 300 v.2010 software.
Serra, K. V., 1988, Well-testing for solution gas drive reservoirs, Ph.D.thesis, University of Tulsa.
Tabatabaie, S. H., 2014, Unconventional Reservoirs: Mathematical Modeling of some Non-linear
Problems, Ph.D. thesis, University of Calgary.
SPE-172928-MS 15

Wattenbarger, R.A., El-Banbi, A.H., Villegas, M.E., and Maggard, J.B. 1998. Production Analysis of
Linear Flow Into Fractured Tight Gas Wells. Paper SPE 39931 presented at the SPE Rocky
Mountain Regional/Low-Permeability Reservoirs Symposium, Denver, 5– 8 April. http://dx.do-
i.org/10.2118/39931-MS.
Whitson, C. and Sunjerga S. 2012. PVT in Liquid-Rich Shale Reservoirs. Paper SPE 155499 presented
at the Annual Technical Conference and Exhibition, San Antonio, Texas, USA. 8-10 October.
http://dx.doi.org/10.2118/155499-MS.
16 SPE-172928-MS

Appendix A
Average properties in the Distance of Investigation (DOI)

For radial flow problems, several investigators (Aanonsen, 1985; Serra, 1988;) have shown that, for constant-rate drawdown
in a homogeneous, infinite-acting gas condensate or solution-gas-drive reservoir, pressures, saturations and any other reservoir
property (␶) are unique functions of the Boltzmann variable i.e. ␶ ⫽f(x/公t). This equation can be arranged to take the following
form:
(A-1)

where ␶ is the dynamic value of any reservoir property at any time and distance. We see that at any instant, different
properties (e.g. pressure, saturation etc.) advance in the porous media using an expression of the form of Equation A-1.
 
If ⌫ represents the volumetric average of ⌫ in the region of influence, and xinv. is the distance of investigation, then ⌫ is
defined as:
(A-2)

With constant values of porosity and thickness, the equation simplifies to:
(A-3)

From basic mathematics we know that ⌫dx ⫽ d(x⌫) ⫺ xd⌫. Equation A-3 therefore becomes:
(A-4)

Which simplifies to:


(A-5)


⌫ in Equation A-5 is independent of time. The conclusion from Equation A-5 is that, when Boltzmann transformation is
valid, the average value of any property in the distance of investigation is invariant with time. Tabatabaie (2014) gives the same
results by transferring the governing partial differential equations from the x-t domain to Boltzmann transformation domain.
SPE-172928-MS 17

Appendix B
Saturation Pressure Relationship in CCE Test

In this section we derive an expression for the volume fraction of oil as a function of pressure in a CCE test. By definition,
the volume fraction of oil is:
(B-1)

where vfo is the volume of free-oil phase and vfg is volume of free-gas phase. Vfo⫽NfoBo and vfg⫽GfgBg where Nfo is
surface volume of oil in the free-oil phase and Gfg is surface volume of gas in the free-gas phase. The total surface volume of
stock tank oil is the summation of surface volume of oil in free gas and oil phases, i. e. N⫽Nfo⫹Nfg. Similarly, the total surface
volume of gas is the sum of gas in free gas and oil phases, i.e. G⫽Gfg⫹Gfo. In these notations, Nfg is the surface volume of
oil in free-gas phase and Gfo is surface volume of gas in the free-oil phase. Nfg and Gfo are functions of PVT properties as
follows: Nfg⫽GfgRV and Gfo⫽NfoRs. By combining these equations and after some mathematical manipulations, oil saturation
can be calculated in terms of PVT properties as given by Eq. 20.
Another way to calculate the saturation change during a CCE experiment is to rely on the fact that Rsi is constant during
any stage of pressure change. This can be stated mathematically as:
(B-2)

With the saturation consistency constraint So⫹Sg⫽1 applied, the saturation equation is then given in Eq. 20:
18 SPE-172928-MS

Appendix C
Material Balance Equation

The generalized material-balance equation can be found in any reservoir engineering textbook (Dake, 1981). This generalized
equation can be rearranged to yield:
(C-1)

Eq. C-1 can be rearranged to the following form if it is assumed that Rv⫽0:
(C-2)

“N” in Eq. C-2 is contacted fluid-in-place, and is a function of time:


Contacted fluid-in-place (t) ⫽ (distance of investigation (t)) ⫻ (2xf) ⫻ (reservoir thickness) ⫻ (porosity) ⫻ (initial fluid
saturation) / (Initial fluid formation volume factor)
The distance of investigation (DOI) calculation for linear geometry is given by Behmanesh et al. 2014 for different
boundary conditions. The general form of DOI is:
(C-3)

There is a substantial increase in two-phase compressibility when gas evolves out of the solution. The total compressibility
at Sg⫽0.05 is assumed for calculation of DOI.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy