Well Test Analysis in Volatile Oil Reservoirs

SPE 116239
Well Test Analysis in Volatile Oil Reservoirs

M. Sanni, SPE, and A.C. Gringarten, SPE, Imperial College London
Copyright 2008, Society of Petroleum Engineers
This paper was prepared for presentation at the 2008 SPE Annual Technical Conference and Exhibition held in Denver, Co, U.S.A., 2124 September 2008.
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
This paper discusses characterization of volatile oil reservoirs using well test analysis. For this purpose, typical well test
behaviors were simulated with a one-dimensional single well compositional reservoir model, for different production rates;
fluid composition and relative permeability curves, with bottomhole pressures above and below the bubble point pressure.
It was found that, when the bottomhole pressure falls below the bubble point pressure during a drawdown, a high gas
saturation zone is created around the wellbore with two-phase (oil and gas) flow, whereas single phase (oil) with the initial gas
saturation remains away from the wellbore. During the subsequent build up, the gas created around the wellbore during the
preceding drawdown condenses into the oil and the saturation in the near-wellbore region returns to the initial gas saturation.
The log-log pressure-derivative behaviors below the bubble point correspond to a two-zone radial composite model, with
decreasing mobility during drawdowns and increasing mobilities during build ups. The log-log pressure derivative plot of the
build up reflects oil mobility distribution of the reservoir at the end of the preceding drawdown.
Knowledge obtained from the study was applied to the analysis of a well test in an actual volatile oil reservoir. Analysis
results were validated with compositional reservoir simulation that included the effect of capillary number and non-Darcy
flow.
Introduction and background

Well tests have been used for many years to assess well condition and obtain reservoir parameters. They remain a very
important component of the reservoir management process for reservoir characterization and evaluation of well performance.
As hydrocarbon exploration moves to deeper geological formations, volatile oil and gas condensate reservoirs have become
increasingly more common. Well tests in volatile oil reservoirs below the bubble point pressure and in gas condensate
reservoirs below the dew point pressure, however, are usually difficult to interpret, because they add the complexity of the
fluid behavior to the complexities of the geology and of the well (Gringarten, 2006). Two important characteristics of fluid
behavior are multiphase flow and capillary number effects.
A better understanding of well test behaviors in volatile oil reservoirs is therefore required to better manage production in
such reservoirs. The present paper uses numerical compositional simulations to identify typical well test behaviors in volatile
oil reservoirs below the bubble point pressure. Results are then applied to actual well test data from a volatile oil field in West
Siberia.
Multiphase well test analysis

Definitions of volatile oil from different authors are summarized in Table 1. Volatile crude oils contain relatively few
heavy hydrocarbon molecules and more intermediate ones compared to black oil. They are also characterized by high liquid
shrinkage immediately below the bubble point (Fig. 1). In a constant temperature depletion experiment, oil viscosity decreases
with decreasing pressure due to the volumetric expansion of oil, until it reaches a minimum value at the bubble point pressure.
Reducing the pressure below the bubble point leads to a net increase in oil viscosity due to liberation of the solution gas until
dead oil viscosity is reached at atmospheric pressure (Khan, 1987).
Below the bubble point pressure, multiphase flow dominates, and fluid relative mobility and reservoir heterogeneities
control the performance (Archer and Wall, 1999). Relative mobility depends on the relative permeability curve, phase
viscosity and average saturation (Cobenas et al., 1999). Muskat and Meres (1963) discussed the equations governing
2 SPE 116239
multiphase flow of fluids through porous media and formulated a basic differential equation for the motion of a heterogeneous
fluid though porous media under steady state and transient conditions. Kniazeff et al. (1965) treated two-phase flow in a
volatile oil reservoir as a binary mixture flow. Two equations of mass continuity based on laws governing composition and
motion were set up and solved numerically.
Published well test analysis methods under multiphase oil-gas conditions use either the Perrine-Martin assumption, or
pseudo-pressure transformations. Perrine (1956) suggested multiphase flow analysis could be done by replacing the single-
phase compressibility and mobility with the sum of compressibilities and mobilities for each phase, respectively. Perrines
method, based on empirical observations, assumes that the different phases are uniformly distributed, with uniform saturations
and permeabilities for each phase. Martin (1959) verified Perrines hypothesis in the case of small saturation gradients.
In the case of a volatile oil below the bubble point pressure, however, there is a high saturation gradient towards the wellbore
and the Perrine-Martin conditions are not met (Ayan et al., 1986, reported that Perrines approach tends to over-estimates the
skin due to gas blockage around the well). Raghavan (1976) proposed a pseudo-pressure approach for wells producing by
solution gas drive, similar to that introduced by Al-Hussainy et al. (1966) for real gas flow:
p
k ro (1)
m( p ) =
p ref
o Bo
dp
The pressure data transformed according to Eq. 1 are then be analyzed to obtain the absolute formation permeability. The
correct definition of the pseudo-pressure integral should actually be:
ko o
P i
(2)
m( p ) =
Pref
dp
o
to obtain a valid Darcy continuity equation:
1 k o o p p (3)
r = c t o
r r o r t
where Darcy flow velocity is defined as:
k p (4)
u= o
o r
Raghavans pseudo-pressure transformation was based on Evinger and Muskats multiphase flow equation for calculating a
theoretical productivity factor. Evinger et al.s solution (1942) assumed steady state flow conditions for heterogeneous fluid
flow in order to simplify their formulation (although the authors acknowledged that such conditions did not exist in practice).
It also used a single permeability-saturation relationship, because of the unavailability of other data at the time of the research.
The authors warned that these simplifications might affect the generalization of their formulation. Al-Khalifah et al. (1987)
developed a diffusivity equation for multiphase well testing in terms of pressures squared:
2 p 2 1 p 2 ct p 2
+ =
r 2 r r t t (5)
which was then linearized using Raghavans pseudo-pressure by assuming a linear relationship between pressure (p) and
(kro/oBo), which is only approximately true (Fetkovich, 1973). Serra et al. (1990a, 1990b) also used flowing wellbore pressure
squared and semilog analysis for estimating effective phase permeabilities and skin effects. This approach is limited to low
producing time and invalid for volatile oil reservoir below the bubble point, due to the continuous increase in gas saturation
around the wellbore.
Capillary number
Multiphase flow in porous media commonly uses the concept of relative permeability functions. When fluid is near-critical,
the multiphase flow relative permeabilities depend on interfacial tension and superficial velocity (Blom et al., 2000). Bardon et
al. (1980) considered the impact of very low interfacial tension on relative permeability. They found that during displacement
of gas in an oil-bearing formation, multiple transfers occur between the liquid and vapor phases so that complete miscibility
may be achieved. As this occurs, the interfacial tension between the two phases reduces progressively to zero. Fulcher et al.
(1985) studied the effect of the capillary number, a dimensionless group representing the ratio of viscous to capillary forces, on
two-phase relative permeability curves. They also verified the shift of the relative permeability from immiscibility towards
miscibility as interfacial tension reduces to zero. Blom et al. (2000) re-defined the capillary number (Nc) as:
k (6)
N =
c
.
Their experiments, using methanol/n-hexane to represent a near-critical gas/condensate or gas/volatile oil system, showed that
the near-critical relative permeability depends on the capillary number at the pore scale (Blom et al. 2000). As initial reservoir
conditions in gas condensate and volatile oil reservoirs are often near critical, the physical properties of the oil and gas phases
are very similar and the interfacial tension between the phases is very low. Bloom et al. (1998) suggested that this may have an
important bearing on the multi-phase flow characteristics in the reservoir during production.
In gas condensate reservoirs with bottomhole pressure below the dew point pressure, high capillary numbers, obtained for high
flowrate or low interfacial tension, have been shown to compensate for the productivity losses due to the creation of a
SPE 116239 3
condensate bank around the wellbore (Gondouin et al. 1967). There has been no study, however, on the effect of the capillary
number in volatile oil reservoirs.
A number of published papers show that capillary number effects may exist with two-phase hydrocarbons other than gas
condensate and volatile oil reservoirs. Talabi et al. (2003) examined the effect of depletion rate and oil viscosity on gas
mobility during solution-gas drive in three viscous oils. They showed that relative permeability to gas was a function of both
gas saturation and oil viscosity; and that gas mobility was low and decreased as oil viscosity or depletion rate (combined in a
depletion index) increased. They did not consider the interaction between rate and viscosity. Ostos et al. (2004) conducted an
experiment to investigate the capillary number effect in heavy oil solution gas drive and its relationship to gas-oil relative
permeability. They showed that the oil produced was a unique function of the capillary number and observed no additional
improvement beyond a critical value. They also found that the oil relative permeability increases and gas relative permeability
decreases with increasing capillary number. Their research suggested that gas-oil relative permeability correlations should
incorporate the effect of capillary number in order to predict production in heavy oil solution gas drive. Bardon et al. (1980)
found that a reduction in interfacial tension increases the oil relative permeability at constant gas saturation in an oil-gas
drainage cycle of the Fontainebleau formation. Rowlison et al. (1982) described similarities in thermodynamics of near-critical
fluids, hence creating a strong basis to investigate effect of capillary number in volatile oil reservoirs.
Simulation model for well tests in volatile oil reservoirs

In the present work, reservoir simulations were performed to generate well test data in a volatile oil reservoir with the
bottomhole pressure above and below the bubble point pressure. Two fluids, A and B, with different properties and
compositions were used in the simulations (Tables 2 to 4). Data for the simulation model are summarized in Table 5.
Grid description
A one-layer radial simulation model was built with 40 cells (Fig. 2). The cell widths increase logarithmically in the radial
direction, with finer grid cells around the wellbore and larger grids further away, in order to capture pressure and fluid
behaviors around the wellbore. The large outer radius of 12000ft ensures that outer boundary effects are not felt during the
simulated well tests.
Fluid characterisation and modeling

The Modified Peng-Robinson equation of state (EOS) with 3 parameters was used for modeling PVT properties of the
reservoir fluids1. Regression was performed on molecular weight (MW) of heavy components; critical pressure (Pc) and
critical temperature (Tc) of the pseudo-components; and binary interaction coefficients between light and heavy components.
Figs. 3 to 8 compare observed and simulated fluid experiments. The Lorentz-Bray-Clark viscosity correlation gives the best
match for fluid sample A, while the Pedersen viscosity correlation gives the best match for fluid sample B. The match error is
less than 10% in each case.
Relative permeability modeling

A Corey type relative permeability model (Blom et al. 2000) :
( Nc )
S S r ( N c ) (6)
k r ( S , N c ) = k * r ( N c )
1 S r ( N c )
was used to generate three different relative permeability curves (Fig. 9). The first relative permeability curve (K1) has a
critical gas saturation (Sgc) of 0.0, an end point (maximum) gas relative permeability (Krgmax) of 0.7, an end point (maximum)
oil relative permeability (Kromax) of 0.95, a connate water saturation (Swc) of 0.15 and a Coreys function exponent () of 2.5.
The second relative permeability curve (K2) was created by reducing the Coreys function exponent from = 2.5 to = 2.0.
This makes the relative permeability curve less steep than the first, hence increasing the relative permeabilities of both oil and
gas phases at a given value of saturation. Krgmax and Kromax were kept at 0.7 and 0.95 respectively. The third relative
permeability (K3) was modeled by reducing the end point (maximum) relative permeability of the oil phase (Kromax) from 0.95
to 0.6 and keeping Krgmax and at 0.7.and 2.5, respectively.
Well test behaviors in volatile oil reservoirs

Simulation runs, consisting of a series of drawdowns (Dd) and build ups (Bu), were designed to investigate the impact on well
test behavior of different reservoir conditions, such as fluid samples and relative permeability models. Simulation runs are
defined by alphanumeric digits indicating the fluid sample (A or B); the relative permeability models (K1, K2, or K3); the
existence of capillary number effect Nc (N0 or N1 for simulations without and with Nc, respectively) and the existence of
turbulence (D0 or D1 for simulations without and with , respectively). Flow periods are labeled DD for drawdown and BU
for build up, preceded by the flow period number. Subscript const-cumm is used for runs with constant cumulative
production. Definition of the flow periods are shown below:
1 DD: constant production rate of 1000 bopd for 3 days
2 BU: shut-in for 3 days
1
Schlumberger: "PVTi Reference Manual version 2001A", 2001
4 SPE 116239

Fig. 10 shows the pressure and rate history for simulation run A-K1-N0-D0 (fluid sample A, relative permeability model 1,
without Nc effect and without effect)
Phase behavior and dynamics

During a well test in a volatile oil reservoir, when the BHP is above bubble point, the saturation and composition are uniform
throughout the reservoir for all DD and BU as shown by the simulated phase profiles in Figs. 11 and 12. A slight increase in
mobility occurs around the wellbore (Fig. 13) corresponding a decrease in oil viscosity due to volumetric expansion (Fig. 14).
When the BHP falls below the bubble point pressure during a drawdown, a high gas saturation zone is created around the
wellbore with two-phase (oil and gas) flow, whereas single phase (oil) with the initial gas saturation remains away from the
wellbore. The size of the near-wellbore high gas saturation region increases with the drawdown duration (Fig. 15). During a
drawdown, the amount of light end components in the fluid decrease towards the wellbore and that of the heavy end
components increase (Fig. 16). During the subsequent build up, the gas created around the wellbore during the preceding
drawdown condenses into the oil and the saturation in the near-wellbore region returns to the initial gas saturation (Fig. 15).
The fluid composition in the near-wellbore region, however, does not return to the initial fluid composition, due to the loss of
the light components (the pressure-temperature phase envelope of the fluid shifts downwards to the right as shown in Fig. 17).
Below the bubble point pressure, the oil viscosity profile depends on the combined effects of the liberation of the solution gas
and the volumetric expansion of the oil. For 6BU (which follows 5DD, a drawdown below the bubble point pressure with a
rate of 2700 Bbl/D), the viscosity of the oil phase increases towards the well in the near-wellbore region. In this build up, the
increase in oil density around the wellbore due to the change in composition over-compensates for the decrease in oil density
due to volumetric expansion (Fig. 18). 4BU, on the other hand, shows a lower oil viscosity near the wellbore because the
decrease in viscosity due to volumetric expansion dominates the increase in oil viscosity due to the change in composition
(Fig. 19). 4BU follows 3DD, a drawdown below the bubble point pressure with a rate of 2300 Bbl/D.
The decrease in oil relative mobility around the wellbore, in all the drawdowns below the bubble point pressure, is due to the
combination of an increase in gas saturation and an increase in oil viscosity. At the end of all the build ups, the gas saturation
in the near-wellbore region returns to its initial value. The oil relative mobility increases around the wellbore at the end of a
build up due to volumetric expansion of fluid. It decreases only when the flow rate of the oil phase is high enough to change
the composition (Fig. 20). Gas becomes mobile as soon as the reservoir pressure falls below the bubble point pressure. The
mobility of the gas phase increases with time during a drawdown and returns to zero at the end of the subsequent build up
when gas re-dissolves into oil (Fig. 21).
Well test behavior of volatile oil reservoirs

The 7% increase in the relative mobility of the volatile oil shown in Fig. 13, when BHP is above the bubble point pressure, is
not significant enough to be seen on the log-log pressure-derivative (1DD and 2BU in Fig. 22). Therefore, the mobility of
volatile oils above the bubble point pressure can be evaluated as for black oil.
Below the bubble point pressure, lower oil relative mobilities can be seen at the beginning of 4BU and 6BU (higher early time
derivative stabilizations in Fig. 23). They correspond to the lower mobilities at the end of 3DD and 5DD (higher late time
derivative stabilizations in Figs. 24 and 25, respectively). The early time derivative stabilizations during the build ups are
actually slightly lower than that the late time derivative stabilizations in the corresponding previous drawdowns because of gas
re-dissolution in oil during the build up, which reduces the oil viscosity and increases the mobility.
The log-log pressure-derivative behaviors below the bubble point therefore correspond to a two-zone radial composite model,
with decreasing mobility during drawdowns (left hand side plots in Figs. 24 and 25) and increasing mobilities during build ups
(right hand side plots in Figs. 24 and 25, and Fig. 26).
Effect of production rate

The impact of the production rate on the well test behavior of volatile oil wells was studied in two ways: (1) increasing oil
production rates with constant flow durations; and (2) increasing oil production rate with constant cumulative production.
Increasing oil flow rates with constant flow durations leads to a decrease in oil relative mobility (Fig. 27). This is due to the
increase in the size of the near-wellbore, high gas saturation region (Fig. 28) and the increase in viscosity towards the well
(Fig. 29). This decrease in oil relative mobility can be clearly seen on the derivative plot of Fig. 30. The same behavior is
obtained when increasing oil production rate with constant cumulative production (11,500 stb in Fig. 31). The increase in the
size of the high gas saturation region is due to the high energy associated with high velocity, which leads to more gas
production than at lower oil rate.
SPE 116239 5
Effect of fluid composition

The highly volatile oil (fluid A) exhibits a higher mobility than the less volatile oil (fluid B) above the bubble point pressure
throughout the reservoir because highly volatile oils are richer in light hydrocarbon components, and therefore have lower
viscosities than less volatile oils. Below the bubble point pressure, the more volatile oil (fluid A) creates higher gas saturations
around the wellbore than the less volatile oil (Fig. 32). Although fluid A has a lower viscosity away from the well (Fig. 33),
the higher gas saturation around the well reduces the oil relative mobility more than for the less volatile oil (fluid B) as shown
in Fig. 34. This behavior can be seen on the log-log pressure derivative plot of Fig. 35, which shows oil mobility being
dominated by saturation in the near-wellbore region and by oil viscosity away from the well.
Effect of relative permeability

The three relative permeability curves shown in Fig. 9 were used to study the effects of relative permeability. Reducing the
steepness (i.e., reducing ) of the relative permeability curves increases Kro at a given saturation (Fig. 36), decreases the
pressure drop around the wellbore (Fig. 37) and increases the oil mobility in the two-phase region of the pressure derivative
(Fig. 38). The reason is that a change in steepness affects the two-phase region more than the end point relative permeability
values, as shown in Fig. 9.
Reducing Kromax decreases Kro throughout the entire reservoir in both the single phase and the two-phase regions (Figs. 39 and
40) and increases the bottomhole pressure drop (Fig. 41).
Effect of capillary number and non-Darcy flow

These effects are discussed below in the context of the actual field data.
Well test analysis of volatile oil reservoir

The well test interpretation model for build ups in a volatile oil reservoir is a two-region radial composite model below the
bubble point pressure, and a homogenous behavior model above the bubble point pressure. A two-region radial composite
reservoir model is characterized by a change in mobility and storativity in the radial direction (Fig. 42), resulting in two radial
flow stabilization on the derivative. Analysis using the composite reservoir model yields the effective permeabilities of the
inner (Keff1) and outer (Keff2) zones; the wellbore skin (S); the total skin (St) due to inner zone; the mobility ratio (kh/)1/2; and
the distance to the radial mobility discontinuity, r1. The actual r1 value is a function of the storativity ratio (cth)1/2, which
requires to calculate the total compressibility in the two-phase region around the wellbore (Ayan and Lee, 1986):
S g dBg dR Bo Rs Bg S dBo dR Bg Rv Bo (7)
Ct = (1 S w ) + v + o + s + S wC w + C f
dP 1 R R B dP dP 1 R R
B g dP s v

o s v

The matches between the build ups from Fig. 10 and the applicable interpretation models are shown in Figs. 43 to 45. Each
build up had to be matched separately because the fluid composition is different in the different flow periods. Analysis results
are summarized in Table 6.
Characterization of volatile oil reservoirs using well test analysis

This section illustrates the application of the findings described above to the analysis of an actual well test (DST-Well-15)
from a volatile oil reservoir in the Pur-Taz region of Western Siberia. Conventional analysis techniques and deconvolution
were supplemented by compositional simulation to verify the existence and the size of the high gas saturation region around
the wellbore, and investigate the impact of the capillary number and non-Darcy flow.
Interpretation of well test data

The pressure and rate histories of DST-Well-15 are shown in Fig. 46. There is an initial drawdown of 41.8 days; a build up of
of 6.4 days (2BU); a drawdown of 26 hours at 3 different rates (4 to 6DD); a build up of 2.7days (7BU); a drawdown of 26
hours at 7 different rates (8 to 15DD); and a final build up of 6.2 days (16BU). The reservoir fluid was initially thought to be
gas condensate, with gas as the dominant fluid and only the measured gas rate was made available. However, fluid analysis
showed the reservoir fluid to be volatile oil (identified as sample B in Tables 2 and 4). The oil rate had to be estimated from
the measured gas rate and the GOR at known separator conditions (Table 7). The oil rate was validated with compositional
simulation and deconvolution. The initial pressure of 4076 psia was obtained from the fluid analysis report (and confirmed by
both conventional analysis and deconvolution).
The analysis was performed in terms of pressure. A rate-normalized log-log plot of all the build ups (Fig. 47) shows a lower
oil mobility at early times (in the two-phase region with high gas saturation), and a higher oil mobility at late times,
corresponding to the reservoir effective permeability. Conversely, the rate normalized log-log plot of all the drawdowns in Fig.
48 shows a higher oil mobility at early times, and a lower mobility at late times. Combination of drawdowns and build ups in
Fig. 49 confirms that the mobility at the end of the drawdowns corresponds to that at the beginning of the build ups, as
predicted by the simulation study (Figs. 24 and 25). Another difference between drawdowns and build ups is apparent in Figs.
47 to 49. Early time drawdown data exhibit non-Darcy flow effects (creating a skin effect that increases with the oil rate).
Early time build up data, on the other hand, show increasing wellbore storage due to phase redistribution in the wellbore (oil is
even re-injected into the reservoir during 16BU, as evidence by a hump and a discontinuity in the corresponding pressure and
derivative, respectively).
6 SPE 116239
No boundary effects are seen on the log-log plots of Figs. 47 to 49, or on the superposition plots in Fig. 50, because the
various build ups were too short (6.4 days maximum). Deconvolution (von Schroeter et al. 2004) was therefore used to
extend the amount of interpretable data, as deconvolution transforms variable rate pressure data into a constant rate initial
drawdown with a duration equal to the total duration of the test (38 days). Fig. 51 shows a log-log plot of the unit-rate
derivative obtained by deconvolving the last build up, 16BU. The deconvolved derivative is identified by a label which
describes the conditions of the deconvolution: (..) refers to the rate record used; [..] identifies the pressure data that have
been deconvolved;{.} states the value of the regularisation parameter (1.54660E+05 in this case); and the last parameter in
the label represents the initial pressure (pav)i, either calculated by deconvolution or imposed to the deconvolution. The initial
pressure can only be calculated by deconvolution if reliable DST data are available at the start of the life of the well.
Alternatively, it can be found by trial and error provided that the test data include at least two build ups: the deconvolved
derivatives for the two build ups are identical at late times if the initial pressure is correct. Otherwise, they diverge or cross
(Levitan, 2005). This procedure confirmed the value of 4076 psia from the fluid analysis report.
The deconvolved derivative in Fig. 51 exhibits a half-unit log-log straight line at late times, suggesting a channel boundary.
The most consistent well test interpretation model is actually a well with wellbore storage and skin in an open rectangular
reservoir (i.e., a channel bounded on one side) with a two-region radial composite behavior. Figs. 52-54 show the match
between data and model, for each build up, on a log-log plot of pressure and derivative, a superposition plot and a pressure
history plot. The match is satisfactory, even though increasing wellbore storage was not included in the model. A skin effect
vs. rate plot showing non-Darcy flow is also included. Well test analysis results are listed in Table 8.
Compositional simulation of DST-Well-15

A compositional simulation of DST-Well-15 was carried out to verify the results of the conventional well test analysis, and, in
particular, the existence and the size of the high gas saturation zone around the well, and the existence and magnitude of non-
Darcy flow. A further objective was to investigate the existence of capillary number effects. The simulation was performed
with Schlumberger Eclipse 300 compositional reservoir simulator and PVTi fluid characterization package. Simulation
parameters are listed in Table 9.
Grid model
A Cartesian grid (52 x 10 x 1) with local grid refinement was used. The simulation model has three sealing faults. Distances
from the well to the faults were set at the values obtained in well test analysis (Fig. 55).
Relative permeability model

Relative permeability curves were generated using Coreys model (Fig. 56). Several relative permeability models and end
point values were tested using compositional reservoir simulation to select the ones that provided a good match with all the
build ups.
Velocity dependence of relative permeability

Fig. 57 compares the depth-adjusted pressure from compositional simulation with the actual pressure history from DST-Well-
15. While the build ups do match, the drawdowns exhibit excessive pressure drops, with the simulated pressure in flow period
15DD even dropping to zero. This suggests that the capillary number, which reduces the pressure drop, affects the actual data.
Both capillary number and non-Darcy flow (identified in the well test analysis) were therefore incorporated in the simulation.
The capillary number in Schlumberger Eclipse 300 compositional simulator is based on Hendersons eight parameter model
(Henderson et al. 2000), whereas non-Darcy flow uses Gertsmas relationship (Gerstma 1974) to estimate the parameters
(Eq. 8):
0.005 cm-1 (8)

k 5.5 S 5.5
0.5
Hendersons parameters were selected by trial and error to provide the best match on the simulated pressure history and log-
log pressure and derivative plots (Figs. 58 to 61). They are listed in Table 10, where values from a North Sea gas condensate
reservoir are also given for comparison. Inclusion of velocity dependence in the relative permeability model does improve the
pressure history match as shown in Fig. 58. The match on the build up pressure and derivative data in Figs. 59 to 61 is not
perfect, possibly because of inaccuracies in the PVT properties.
Phase behavior
Fig. 62 shows the oil saturation profiles at the end of all the build ups. Although the gas around the well has dissolved into the
oil, a two-phase region still exists away from the well. This is because, with the initial reservoir pressure at the bubble point
pressure and the long duration of the first drawdown, the build ups are too short for the reservoir pressure to return to the
initial pressure.
Fig. 62 also displays the radius of the high gas saturation region estimated from well test analysis. It compares reasonably well
with that from compositional simulation, although the simulation shows a continuous change in mobility rather than an abrupt
one.
SPE 116239 7
Conclusions
This objective of this paper was to identify typical well test behaviors in volatile oil reservoirs below the bubble point pressure.
It was found that, when the bottomhole pressure falls below the bubble point pressure during a drawdown, a high gas
saturation zone is created around the wellbore with two-phase (oil and gas) flow, whereas single phase (oil) with the initial gas
saturation remains away from the wellbore. During the subsequent build up, the gas created around the wellbore during the
preceding drawdown condenses into the oil and the saturation in the near-wellbore region returns to the initial gas saturation.
The impairment to flow due to the high gas saturation zone around the wellbore when the bottom hole pressure is below the
bubble point pressure can be seen as a mobility contrast in well test analysis. The log-log pressure-derivative behaviors below
the bubble point therefore correspond to a two-zone radial composite model, with decreasing mobility during drawdowns and
increasing mobilities during build ups. The log-log pressure derivative plot of the build up reflects oil mobility distribution of
the reservoir at the end of the preceding drawdown.
High volatility oils have higher mobilities than less volatile oils above bubble point pressure. However, higher gas saturation
below bubble point pressure causes relatively larger mobility reductions for the more volatile oils.
Acknowledgments
This work was supported by a consortium of industrial sponsors comprising the UK Department of Trade and Industry, BG,
ConocoPhillips, Eni, Petrom, PetroSA and Total. We are grateful to Schlumberger (Eclipse 300, PVTi) and Paradigm
(interpret) for providing the software used in this study. Sponsorship of Moshood Sannis PhD studies by Petroleum
Technology Development Fund, Nigeria is gratefully acknowledged.
Nomenclature
a Empirical constant in k o = ap relationship Ng: number of pseudo components
o Bo N/G: net to gross ratio
BHP: Bottom Hole Pressure. P: pressure
Bopd: Barrel of oil per day Pref: reference pressure
B: formation volume factor for phase Pav: average initial reservoir pressure
BU: build up. Pi: initial reservoir pressure
c: compressibility Pbub: bubble Point pressure
ct: total compressibility Pc: critical pressure
DD: drawdown. P-T: pressure temperature.
EOS: equation of Sate. PVT: pressure Volume Temperature
GOR: producing gas to oil ratio Rs: solution gas /oil ratio
h: formation thickness ri: radius of Investigation
k: absolute Permeability r1: radius of composite discontinuity
Keff: effective reservoir permeability rw: wellbore radius
Kr: relative permeability Rv: dissolved oil/gas ratio
K: effective permeability of phase s: skin
Kr: relative permeability of phase S: saturation
Krmax: maximum relative permeability of phase Sgc: critical gas saturation
Kr*: end point relative permeability Sr: residual saturation
m: Slope of the infinite acting semilog straight line Sr: residual saturation of phase
m(p): pseudo-pressure Tc: critical reservoir temperature
MW: molecular weight Tr: reservoir temperature
Nc: capillary number.
Greek : flow potential

: viscosity
v: velocity Subscripts
: porosity : phase
: density g: gas
: factor representing the inertia/turbulence or non-Darcy o: oil
flow effect. t: total
: Corey exponent in the relative permeability function. eff: effective
: mobility ratio of phase defined as (kr/). const-cumm: Constant cumulative production
: interfacial tension
References
1. Archer, J.S. and Wall, C.G.: "Petroleum Engineering Principles and Practice", Kluwer Academic Publishers Group,
Dordrecht / Boston, 1999.
2. Al-Hussainy, R., Ramey Jr., H. J., and Crawford, P. B.: The Flow of Real Gas Through Porous Media, Journal of
Petroleum Technology (May 1966), pp.637-642.
8 SPE 116239
3. Al-Kahlifa, A.J.A., Aziz, K., and Horne, R.N.: "A New Approach to Multiphase Well Test Analysis", paper SPE 16743
presented at the 62nd Annual Technical Conference and Exhibition held in Dallas, Texas, Sept. 27-30, 1987.
4. Ahmed, T.: "Hydrocarbon Phase Behavior", Contributions in Petroleum Geology & Engineering, Gulf Publishing
Company, 1989.
5. Ayan, C. and Lee, W.J.: "Effects of Multiphase Flow on interpretation of Buildup Tests", Society of Petroleum Engineers
Formation Evaluation, (June 1988), pp. 459-466.
6. Bardon, C. and Longeron, D.G.: "Influence of Very Low Interfacial Tensions on Relative Permeability", Society of
Petroleum Engineers Journal (Oct. 1980), pp. 391-401.
7. Blom, S.M.P. and Hagoort, J.: "How to Include Capillary Number in Gas Condensate Relative Permeability Functions",
paper SPE 49628 presented at the SPE Annual Technical Conference and Exhibition held in New Orleans, Louisiana, Sept.
27-30, 1998.
8. Blom, S.M.P., Hagoort, J., and Soetekouw, D.P.N.: "Relative Permeability at Near-Critical Conditions", Society of
Petroleum Engineers Journal (June 2000), pp. 172-181.
9. Coats, K.H. and Smart, G.T.: "Application of Regression-Based EOS PVT Program to Laboratory Data", Society of
Petroleum Engineers Reservoir Engineering (May 1986), pp. 277-299.
10. Cobenas, R.H. and Crotti, M.A.: "Volatile Oil. Determination of Reservoir Fluid Composition from a Non-Representative
Fluid Sample", paper SPE 54005 presented at the 1999 SPE Latin American and Caribbean Petroleum Engineering
Conference held at Caracas, Venezuela, Apr. 21-23, 1999.
11. Evinger, H.H. and Muskat, M.: "Calculation of theoretical productivity Factor", Trans. AIME 146, 1942, pp. 126-139.
12. Fetkovich, M.J.: "The Isochronal testing of oil wells", paper SPE 4529 presented at the 48th Annual fall Meeting of SPE-
AIME held at Las Vegas, Nev., Sept. 30-Oct. 3, 1973.
13. Fulcher, R.A.Jr., Ertekin, T., and Stahl, C.D.: "Effect of Capillary Number and Its Constituents on Two-Phase Relative
Permeability Curves", Journal of Petroleum Technology, (Feb. 1985), pp. 249-260.
14. Geertsma, J.: "Estimating the Coefficient of Inertial Resistance in Fluid Flow Through Porous Media", Journal of
Petroleum Technology (Oct. 1974), pp. 445-450.
15. Gondouin, M., Iffly, R. and Husson, J.: An Attempt to Predict the Time Dependence of Well deliverability in Gas-
Condensate Fields, Society of Petroleum Engineers Journal (June 1967), pp. 112-124.
16. Gringarten A. C., Daungkaew S., Hashemi S. and Bozorgzadeh, M: ''Well Test Analysis in Gas Condensate Reservoirs:
Theory and Practice,'' paper SPE 100993, presented at the 2006 SPE Russian Oil and Gas Technical Conference and
Exhibition, Moscow, Russia, 36 October 2006.
17. Henderson, G.D., Danesh, A., Tehrani, D.H., Al-Kharusi,B.: "Generating Reliable Gas Condensate Relative Permeability
Data Used to Develop a Correlation with Capillary Number," Journal of Petroleum Science and Engineering 25 (2000),
pp. 79-91.
18. Khan, S.A.: "Development of Viscosity Correlation for Crude Oils", Society of Petroleum Engineers Journal (Aug 1987)
19. Kniazeff, V.J. and Naville, S.A.: "Two-Phase Flow of Volatile Hydrocarbons", Society of Petroleum Engineers Journal,
March 1965, pp. 37.
20. Levitan, M. M.: Practical Application of Pressure/Rate Deconvolution to Analysis of Real Well Tests, Society of
Petroleum Engineers Reservoir Evaluation & Engineering (April 2005) pp. 113-121.
21. Martin, J.C.: "Simplified Equations of Flow in gas Drive reservoirs and the Theoretical Foundation of Multiphase Pressure
Buildup Analyses", Trans., AIME 216 (March 1959) pp. 321-323.
22. McCain, W.D.J.: "Properties of Petroleum Fluids", Penwell Books, Penwell Publishing Co, 1990.
23. Moses, P.L.: "Engineering Application of Phase Behavior of Crude Oil and Condensate System", Journal of Petroleum
Technology (July 1986), pp. 715-723.
24. Muskat, M. and Meres, M.W.: "The flow of heterogeneous fluids through porous media", Journal of Applied Physics 7
(Jul. 1963), pp. 346-363.
25. Ostos, A. and Maini, B.: "Capillary Number in Heavy Oil Solution Gas Drive and Its Relationship with Gas-Oil Relative
Permeability Curves", paper SPE 89430 presented at the 2004 SPE/DOE 14th Symposium on Improved Oil recovery held
in Tulsa, Oklahoma, U.S.A, Apr. 17-21 2004.
26. Perrine, R.L.: "Analysis of pressure Build up curves", API, Dallas, Drill & Prod. Prac, Jan. 1956, pp. 482-509.
27. Raghavan, R.: "Well test Analysis: Wells Producing by Solution Gas-Drive", Society of Petroleum Engineers Journal
(1976) 196-208; Trans., AIME.
28. Rowlinson, J.S. and Widom, B.: "Molecular Theory of Capillarity", Courier Dover Publications, Inc, 1982.
29. Serra, K.V., Peres, A.M.M., and Reynolds, A.C.: "Well test Analysis for Solution-gas-Drive Reservoirs: Part 2 Buildup
Analysis", Society of Petroleum Engineers Formation Evaluation, June, 1990a.
30. Serra, K.V., Peres, A.M.M., and Reynolds, A.C.: "Well test Analysis for Solution-gas-Drive Reservoirs: Part3 - A unified
treatment of the Pressure-Squared Method", Society of Petroleum Engineers Formation Evaluation, June, 1990b.
31. Talabi, O., Pooladi-Darvish, M., and Okazawa, T.: "Effect of Rate and Viscosity on Gas Mobility during Solution-Gas
Drive in Heavy Oils", paper SPE 84032 presented at the SPE Annual Technical Conference and Exhibition held Denver,
Colorado, USA, Oct. 5-8 2003.
SPE 116239 9
32. Von Schroeter, T., Hollaender, F., and Gringarten, A.C.: "Deconvolution of Well Test Data as a Nonlinear Total Least-
Square Problem", Society of Petroleum Engineers Journal (Dec. 2004), pp. 375-390.
33. Whitson, C.H. and Brul M.R.: "Phase Behavior", SPE Monograph, Henry L.Doherty Series 20 (Jan. 2000).
Table 3 Composition of fluid sample A

Components Fraction
Table 1 Literature definitions of volatile oil CO2 0.0090
Author Definition API GOR Bo
N2 0.0030
(scf/stb) (rb/stb)
Whitson et High GOR, 35o 1,000 - > 1.5 C1 0.5347
al. 2000 High 3,000 C2 0.1146
shrinkage to C3 0.0879
50% C4 0.0456
Moses1986 High 40o 2,000 - 2.0
shrinkage 3,500 C5 0.0209
immediately C6 0.0151
below the C7+ 0.1692
bubble point MW+ 173
pressure.
High + 0.83648
shrinkage
(can go as
high as 45%)
Ahmed Produce more 45o to 2,000 - 2.0 Table 4 Composition of fluid sample B
1989 gas than black 55o 3,500
Composition Fraction
oil for same
pressure drop. N2 0.0087
Greenish to
Co2 0.0016
orange colour
McCain, Relatively 40o 2,000 - 2.0 H2S 0.0000
1990 fewer heavy 3,300 C1 0.4943
molecules and
more C2 0.0728
intermediates. C3 0.0802
Brown,
Orange or IC4 0.0231
green colour NC4 0.0361
IC5 0.0180
Table 2 Summary of fluid properties NC5 0.0179
Fluid Samples C6 0.0232
A B C7+ 0.2241
Very Volatile
Fluid Type Oil Moderately Volatile Oil
Pbub ( psia) o
4,475 at 176 F 4,076 at 189oF
SG @ 60 oF 0.8479
o
Rs (scf/bbl ) 3,377 at 176oF 1,786 at 189oF API @60oF 35.4
Bo( bbl/stb) MW 215.0
at Pbub 2.92 2.02
Coats and
Source Smart 1982 Western Siberia
10 SPE 116239
Table 5 Reservoir model characteristics Table 9 Parameters for simulation of well test DST-Well-15
Sample Parameter Value
Parameter A B Wellbore Radius (ft) 0.31
Reservoir Thickness (ft) 20
Porosity , % 15 15 Net to Gross Ratio 1
Absolute permeability k, mD 10 10 Top of Reservoir (ft) 10,000
Net to-Gross ratio N/G 1 1 Initial Reservoir Pressure at 4067
Wellbore radius rw, ft 0.2 0.2 gauge Depth (psia)
Top depth, ft 10,000 10,000 Gauge Depth (ft) 9810 (190 ft above top perforation)
Reservoir thickness, ft 100 100 Average Radial Reservoir 19 mD
Reservoir temperature, oF 176 189 Permeability (mD)
Initial reservoir pressure Pi 5260 4861 Kv/Kh Ratio 0.1
Reservoir Porosity (%) 20%
Pi Pbub, psia 785 785 Initial water Saturation (Swi) 0.4
Bubble Point Pressure (psia) 4067
Table 6 Result of well test analysis d1 (ft) 5165
parameters Analysis of Build ups d2 (ft) 2712
d3 (ft) 1190
2BU 4BU 6BU
Analysis Homoge Radial Radial
Model nous composite composite
Pi (psia) 5260 5260 5260 5260
Table 10 Input Parameter for Velocity-Dependent Relative
Koeff (K x 9.5 9.5 9.5 9.5 Permeability Model
Kromax) Value for
K (mD) 10 10 10 10 Values for North Sea
S(t) 0 0.0 3 5 Parameter Description
Simulation Gas
Ri (ft) NA 936 726 640 Condensate
(ch)1/2 NA NA 2.5 3.2 mg Controls the variability 10 23.89
(from of the critical gas
fluid saturation with the
analysis) normalised capillary
number
(Kh/)1/2 NA NA 0.67 0.54 mo Controls the variability 0 79.62
of the critical oil
Table 7 Estimation of oil rate from measured gas rate and saturation with the
GOR-DST-Well-15 normalised capillary
Flow Period Recorded Gas Estimated Oil Rate Using number
Rate measured GOR n1g Controls the weighting 5 6.23
Mscf/D Stb/D between the miscible
1 DD 1000 560 and immiscible relative
2 BU 0 0 permeability curves
3DD 0 0 (along with n2g
4DD 1000 560 parameter)
5DD 770 431 n2g Controls the weighting -1 0
6DD 700 392 between the miscible
7BU 0 0 and immiscible relative
8DD 0 0 permeability curves
9DD 225 126 (along with n1g
10 DD 860 482 parameter)
11DD 960 538 n1o Controls the weighting 0.5 24.2
12DD 890 498 between the miscible
13DD 1350 756 and immiscible relative
14DD 1525 854 permeability curves
15DD 1530 857 (along with n2o
16BU 0 0 parameter)
n2o controls the weighting 0 0
between the miscible
Table 8 Summary of well test analysis of DST-Well-15 using and immiscible relative
Build up data permeability curves
Variatio (along with n1o
Parameter Flow Periods n% parameter)
16B Ncbo Base capillary number 1.0E-06 1.0E-06
2BU 7BU U for oil. This is the
(pav)i 4076 4076 4076 psia 0 threshold value of the
pwf 2529 2996 2175 psia - capillary number above
which the VDRP effect
kh(2) 380 380 410 mD.ft 5 is thought to be active.
k2 eff 19 19 20 mD 5 Ncbg.n Base capillary number 1.0E-011 1.0E-06
C 0.05 0.01 0.04 bbl/psi - for gas. This is the
S(t) 1.9 1.35 2.7 - threshold value of
r1 260 260 260 ft 0 capillary number above
Fluid which the VDRP effect
(pch)1/2 2.1 1.6 2.6 Analysis - is thought to be active
(kh/u)1/2 0.3 0.2 0.3 -
d1 5160 5160 5160 ft 0
d2 2710 2710 2710 ft 0
d3 1190 1190 1190 ft 0
SPE 116239 11
100% Bubble
point
Low shrinkage crude oil DL1 Oil rel vol
2.75 Calculated
Observed
O il re lative vo lum e (R B /S TB )
2.50
Liquid Volume
Residual oil 1.25

High shrinkage below
50% bubble point leading to
significant decrease 2.00
(about 50%) immediately
Near critical point crude oil
below bubble point 1.75
Residual oil 1.50
1.25
1.00
0% 4000
0 1000 2000 3000
Pressure (psia)
Pressure (psig)
Figure 1 Shrinkage curve for crude oil system (Ahmed, 1989) Figure 5 DL for Oil Relative Vol. Fluid A
CCE Liquid Viscosity

Calculated
1.25
Observed
intrad=0.2 ft
Liquid Viscosity (CP)

1.00
0.75
K=10mD =15% 100ft
0.50
Homogenous Reservoir
0.25
12000ft
1000 2000 3000 4000 5000
Pressure (psig)
Figure 2 Radial simulation grid model (40 x 1 x 1) Figure 6 CCE for Oil Viscosity Fluid B
DL1 Liquid Density

54
CCE1 Relative Vol Calculated
52 Observed
Calculated
Liquid density (lb/ ft3)
Observed 50
Oil Relative Volume
48
2.00 46
44
42
40
1.00
5000 38
1000 2000 3000 4000 1000 2000 3000 5000
4000
Pressure (psig)
Pressure (psig)
Figure 3 CCE for Relative Volume Fluid A Figure 7 DL for Liquid Density Fluid B
DL1 Gas Gravity

CVD1 Moles Recovered
2.00
0.60 Calculated
Observed
Calculated 1.75
0.50
Moles Recovered
Gas Gravity
Observed
0.40 1.50
0.30 1.25
0.20 1.00
0.10
1000 2000 3000 4000 1000 2000 3000 4000
Pressure (psig) Pressure (psig)

Figure 8 DL for Gas Gravity Fluid B
Figure 4 CVD for Moles Recover Fluid A
12 SPE 116239
1 0.175
Initial reservoir viscosity
0.9
0.8 Model 1: Kromax=0.95, Krgmax=0.7 =2.5 0.170
0.7 7% Decrease
Oil Viscosity (cp)

0.6 Model 2: Kromax=0.95, Krgmax=0.7 =2.0 0.165
K ro/K rg
0.5
0.4 0.160
1DD 2BU
0.3 Model 3:
Kromax=0.60,
0.2 Krgmax=0.7 =2.5 0.155
0.1
0 0.150
10 -1 10 0 10 1 10 2 10 3 10 4
0 0.2 0.4 0.6 0.8 1
Radial Distance from Wellbore (ft)
Sg
Figure 13 Effect of volumetric expansion on oil mobility:
Figure 9 Coreys model relative permeability run A-K1-N0-D0
6000 6000 6.0
5000 5.9
5000 1DD 2BU
Pbub
Oil Relative Mobility (cp-1)

Oil Rate (STB/D)
psia 4000 5.8

Pressure (psia)
4000
3000 5.7
7% increase in oil mobility
3000
2300 2700
2000 5.6
1000 STB/D STB/D
STB/D
2000
1000 5.5
0 0 0 Initial reservoir relative mobility
STB/D STB/D STB/D
1000 0
5.4
0 2 4 6 8 10 12 14 16 18 20 22 24 26
10 -1 100 10 1 10 2 10 3 10 4
Elapsed time (days) Radial Distance from Wellbore (ft)
Figure 10 Pressure-rate history for simulation: Figure 14 Effect of volumetric expansion on oil viscosity:
run A-K1-N0-D0 run A-K1-N0-D0
1 1
0.9
Initial reservoir saturation
0.8 0.9
Initial reservoir saturation
0.7
0.6 0.8
Oil Saturation
0.5
1DD 2BU 0.5 days of 5DD
Oil Saturation
0.7 End of 5DD

0.4
0.0001 days of 6BU
0.3
End of 6BU
0.6
0.2
0.1
0.5
0
10 -1 10 0 10 1 102 10 3 10 4
0.4
10 -1 10 0 10 1 10 2 10 3 10 4
Figure 11 Constant saturation for flow periods above Pbub Radial Distance from Wellbore (ft)
: run A-K1-N0-D0 Figure 15 Saturation profile for Flow Periods below Pbub :
run A-K1-N0-D0
0.6
0.6
3DD
0.5
0.5
Fractional Composition
CO2 N2 C1 C2 5DD
Fractional Composition
0.4 CO2 N2 C1 C2 C3
0.4
C3 C4 C5 C6 C7+
C4 C5 C6 C7+
0.3 0.3
5DD
0.2 0.2
3DD
0.1 0.1
0 0
10 -1 10 0 10 1 10 2 10 3 10 4
10 -1 10 0 10 1 10 2 10 3 10 4
Radial Distance from Wellbore (ft) Radial Distance from Wellbore (ft)
Figure 12 Constant composition for flow periods above Figure 16 Composition profile for Flow Periods below
Pbub : run A-K1-N0-D0 Pbub : run A-K1-N0-D0
SPE 116239 13
6000 0.7
Beginning of 5DD
Reservoir depletion path 176 oF End of 5DD
5000 0.6
End of 6BU
Gas Relative Mobility (cp-1)

0.5
Pressure (psia)
4000
Change in phase
envelope at end 5DD
0.4
of 6BU.
3000
0.3
2000
0.2 3DD
1000
0.1
0 4BU 6BU
0 100 200 300 400 500 600 700
0
10 -1 10 0 10 1 10 2 10 3 10 4
Temperature (oF)
Figure 17 Change in phase in envelope due to loss of light
end hydrocarbon: run A-K1-N0-D0 Figure 21 Gas relative mobility profile at end of flow
periods; run A-K1-N0-D0
0.24
S<Soi S=Soi S=Soi

0.22
<oi <oi =oi 103 103
P res su re C han ge a nd
0.2
D e riva tive (psia )

Oil Viscosity (cp)
2BU
0.18 1DD
0.16 102 102
0.14 Volumetric expansion

Increase in o
0.12 5DD 6BU
101 101
0.1 10-5 10-4 10-3 10-2 10-1 100 101 102 10-5 10-4 10-3 10-2 10-1 100 101 102
10 -1 10 0 10 1 10 2 10 3 10 4
Elapsed time (hrs) Elapsed time (hrs)
Figure 18 Oil viscosity profile at end of 5DD and 6BU: Figure 22 Log-Log Pressure Derivative plot for flow
run A-K1-N0-D0 above bubble point: run A-K1-N0-D0
0.24
0.22 10 3
2BU
4BU
Change and Derivative (psi)
3DD 4BU
Rate Normalised Pressure
0.20 6BU
Oil Viscosity (cp)
0.18
0.16
10 2
Low mobility which can corresponding only
0.14 to end of preceding draw down (3DD and 5DD)
0.12 Decrease in o due to volumetric expansion
0.10
10 0 3 4
10 -1 10 1 10 2 10 10 10 1
10-5 10-4 10 -3 10 -2 10 -1 10 0 101 10 2
Figure 19 Oil viscosity profile at end of 3DD and 6BU: Elapsed time (hrs)
run A-K1-N0-D0 Figure 23 Log-Log pressure derivative showing

decreasing oil mobility towards wellbore: run A-K1-N0-
D0
7
6
Initial Mobility
104 104
P re s su re C h a n g e a n d
5
4BU
Oil Relative Mobility (cp-1)
3DD
D e riva tive (p s ia )
Higher Mobility Lower Mobility

4 103 103
3DD 5DD
3
Mobility around well
End of 3DD and 5DD More 4BU 6BU
102 102
2 than 50% mobility change
Original mobility around well

1
101 101
10-5 10-4 10-3 10-2 10-1 100 101 102 103 10-5 10-4 10-3 10-2 10-1 100 101 102 103
0
10 -1 10 0 10 1 10 2 10 3 10 4 Elapsed time (hrs) Elapsed time (hrs)
Figure 24 Mobility at end of 3DD corresponding with
Figure 20 Oil Relative Mobility profile at end of 5DD and
mobility at the beginning of 4BU: run A-K1-N0-D0
6BU: run A-K1-N0-D0
14 SPE 116239
0.25
104 104
P re s su re C h a n g e a n d
Oil Viscosity (cp)

0.2
6BU
5DD
D e riva tive (p s ia )
103 103 0.15
Mobility around well Increasing viscosity around wellbore

0.1
102 102
5DD with Qo =2300 STB/D
Original mobility around well
0.05 5DD with Qo =2700 STB/D
101 101
10-5 10-4 10-3 10-2 10-1 100 101 102 103 10-5 10-4 10-3 10-2 10-1 100 101 102 10
3
0
Elapsed time (hrs) Elapsed time (hrs) 10-1 100 101 102 103 104
Figure 25 Mobility at end of 5DD corresponding with Figure 29 Oil viscosity profile: run A-K1-N0-D0
mobility at the beginning of 6BU: run A-K1-N0-D0
104
104
Qo =-2300 STB/D
Change and Derivative (psia)

Qo =-2700 STB/D
Pressure Change and Derivative (psia)

103
103
Lower relative mobility
K eff K eff
Mobility around well <
o 2 o 1 102
K eff
102
o 1
Initial reservoir mobility (mobility further into reservoir)
Early decrease in relative mobility
10
101 10-5 10-4 10-3 10-2 10-1 100 101 102 103
10-5 10-4 10-3 10-2 10-1 100 101 102 103 Elapsed time (hrs)
Elapsed time (hrs) Figure 30 Log-Log derivative plot showing effect of
production: run A-K1-N0-D0
Figure 26 Two-zone composite behavior of volatile oil
reservoir below bubble point
103
7
Early decrease in relative mobility
Cumulative production of oil =11500 STB for each flow period

6
O il R e la tive M o b ility (cp -1 )
5 102
4 Lower mobility
3 5DD with Qo =2700 STB/D

101
Duration of 5Days
2 Keff

Q = -2300stb/d o 1
1 Lower relative mobility
Q = -2700stb/d
10
0 10-5 10-4 10-3 10-2 10-1 100 101 102 103
10-1 100 101 102 103 104 Elapsed time (hrs)

Figure 31 Log-Log derivative plot showing effect of
Figure 27 Oil relative mobility profile: run A-K1-N0-D0 production: run A-K1-N0-D0const cumm
1
1
0.9
0.9
0.8
0.8
0.7
0.7
Oil Saturation
Oil Saturation
0.6
0.5
0.6
Increasing gas saturation zone Fluid A
0.4 0.5
5DD with Qo =2300 STB/D Fluid B
0.3 0.4
0.2 0.3
0.2 More volatile (Fluid A) showing higher gas saturation around well
0.1
0.1
0
10-1 100 101 102 103 104 0
10-1 100 101 102 103 104
Figure 28 Oil saturation profile: run A-K1-N0-D0 Figure 32 Oil saturation profile showing effect of fluid
composition: run A-K1-N0-D0and B-K1-N0-D0
SPE 116239 15
5300
0.35 5100
Fluid A 4900
0.30
Fluid B 4700
0.25
Oil Viscosity (cp)
4500
Pressure (psia)
0.20 4300
0.15 4100
3900
0.10
More volatile (Fluid A) showing lower initial and final viscosity 6BU with =2.0 (A-K2-N0-6BU)
3700
0.05
3500 6BU with =2.5 (A-K1 -N0-6BU)
3300
10-1 100 101 102 103 104 0 2 4 6 8 10 12 14 16 18 20 22 24 26
Radial Distance from Wellbore (ft) Elapsed time (days)
Figure 33 Oil viscosity profile showing effect of fluid Figure 37 Pressure history for A-K1-N0-D0 and
composition: run A-K1-N0-D0and B-K1-N0-D0 A-K2-N0-D0
7 104
6BU with =2.0 (A-K2-N0-6BU)
Effect of saturation reducing

6 mobility of more volatile oil 6BU with =2.5 (A-K1-N0-6BU)
Oil Relative Mobility (cp -1 )

at higher rate Fluid A
5
Fluid B 103
4
3
Same initial; mobility
2 Viscosity dominates mobility 102
1
Increase in mobility due to reduction in
0
10-1 100 101 102 103 104 10
10-5 10-4 10-3 10-2 10-1 100 101 102 103
Elapsed time (hrs)
Figure 34 Oil relative mobility profile showing higher Figure 38 Log-log pressure derivative plot showing effect
mobility for more volatile oil further from wellbore: of : run A-K1-N0-D0 and A-K2-N0-D0
run A-K1-N0-D0and B-K1-N0-D0
104 1
Pressure Change and Derivative (psi)
0.9
Oil Relative Perm eability
A-K1-N0-6BU
5DD with K romax = 0.95
0.8 5DD with K romax = 0.60
103 0.7
Fluid A
Effect of saturation 0.6
Fluid B
0.5 Decrease in relative permeability
0.4 of oil far from well due to decrease in Kromax
102
0.3
0.2
Viscosity dominates mobility
0.1
101
0
10-5 10-4 10-3 10-2 10-1 100 101 102 103
10-1 100 101 102 103 104
Elapsed time (hrs) Radial Distance from Wellbore (ft)

Figure 35 Log-log pressure derivative plot showing effect Figure 39 Oil relative permeability profile showing effect
of fluid composition: run A-K1-N0-D0and B-K1-N0-D0 of Kromax: run A-K1-N0-D0 and A-K3-N0-D0
104
Pressure Change and Derivative (psia)
1
0.9 Increasing k ro for
same value of So
Oil Relative Permeability
0.8
103
5DD with K romax = 0.95
0.7 5DD with K romax = 0.60
0.6
0.5
0.4 102
0.3 Decrease in mobility

due to reduction in Kromax
0.2 6BU with =2.0 (A-K2-N0-6BU) Different initial reservoir mobility
0.1
6BU with =2.5 (A-K1-N0-6BU) 101
0 10-5 10-4 10-3 10-2 10-1 100 101 102 103
10-1 100 101 102 103 104 Elapsed time (hrs)
Figure 40 Log-log pressure derivative plot showing effect
Figure 36 Oil relative permeability profile showing effect of Kromax: run A-K1-N0-D0 and A-K3-N0-D0
of : run A-K1-N0-D0 and A-K2-N0-D0
16 SPE 116239
102
6000 5500
Pressure Change and
Pressure (psia)
Derivative (psi)
5000
Pressure change and Simulated pressure
5000 4500 match for 6BU
101 derivative match 6BU
4000
4000 3500
Pressure (psia)
10 3000
500 520 540 560 580 600 620
10-5 10-4 10-3 10-2 10-1 100 101 102
3000 Elapsed time (hrs) Elapsed time (hrs)
5200
2000 5000
Pressure (psia)
4800
4600
1000 5DD with K romax = 0.95 4400
5DD with K romax = 0.60 4200
Horner plot match for 6BU
Increase in P 4000
0 3800
3600
0 2 4 6 8 10 12 14 16 18 20 22 24 26 0 4000 8000 12000 16000
Superposition Function (STB/D)
Elapsed time (days)
Figure 41 Pressure history: run A-K1-N0-D0 and Figure 45 Well test analysis match for 6BU: run
A-K3-N0-D0 A-K1-N0-D0
5000
pbub=pi 2BU 7BU 16BU

well
4000 1800
1600
Oil Rate (STB/D)

1400
Pressure (psia)
3000
(c t h)1 kh
1200
1 1000
2000
800
(c t h)1 > (c t h )2 kh kh
Missing Pressure data for 41 days 600
< 1000
400
1 2
200
0 0
0 10 20 30 40 50 60 70
Figure 42 Radial composite model in a volatile oil Elapsed time (days)
reservoir Figure 46 DST Pressure Rate History of DST-Well-15
104
102 5300
Pressure Change and
2 BU
Pressure (psia)
5200
7 BU
Derivative (psi)
5100
Phase redistribution 16 BU
5000
101 match for 2BU
derivative match 2BU
4900 103
4800
10 4700 kh kh
10-5 10-4 10-3 10-2 10-1 100 101 102 60 70 80 90 100 110 120 130 140 <
1 2
5300 102 kh

5200 1
Pressure (psia)
Corresponding to initial reservoir mobility

5100
5000
4900 Horner plot match for 2BU 10

10-2 10-1 100 101 102 103
4800
0 1000 2000 3000 4000 5000 6000 7000 Elapsed time (hrs)
Figure 47 Combination of all Log-Log Pressure and

Figure 43 Well test analysis match for 2BU: run Derivative Plots for Build up
A-K1-N0-D0
104
102 6000
Pressure Change and
Pressure (psia)
4 DD (559 stb/day)
Derivative (psi)
5000 10 DD (481 stb/day)


101 103 13 DD (755 stb/day)

derivative match 4BU match for 4BU
4000
14 DD (853 stb/day)
10 3000
10-5 10-4 10-3 10-2 10-1 100 101 102 240 260 280 300 320 340 360 380
Elapsed time (hrs) Elapsed time (hrs) 102
5400
Pressure (psia)
5200
5000 Increasing skin
101
4800
4600 Two-phase mobility
4400
4200 Horner plot match for 4BU
4000 100
0 4000 8000 12000 16000
10-2 10-1 100 101
Elapsed time (hrs)
Figure 48 Combinations of Log-Log Pressure and

Figure 44 Well test analysis match for 4BU: run
Derivative Plots for Drawdowns
A-K1-N0-D0
SPE 116239 17
104 4500
and Derivative (psi)

103 Log-Log Match 7BU Simulation 7BU
Pressure (psia)
Pressure Change
4000
103 3500
DD BU
3000
2-phase mobility 102

2500
10 -2 2000
10 10-1 100 101 102 103 1000 1100 1200 1300 1400 1500
102 2.0 3800

Skin Vs. Rate Horner Match 7BU
Pressure (psia)
1.5
3600
Skin
1.0
Initial reservoir mobility 3400
Initial mobility stabilisation 0.5
cannot be seen from the DD 3200
0.0
-0.5 3000
100 300 500 700 900 1100 400 800 1200 1600 2000
Oil Rate (STB/D) Superposition Function (STB/D)
10
10-2 10-1 100 101 102 103
Elapsed time (hrs) Figure 53 Analysis Match on Log-Log Pressure Change

Figure 49 Combinations of log-log pressure and & Derivative; Pressure History Simulation; Horner and
derivative plots for drawdowns and Build ups Skin vs Rate using Open Ended Rectangle boundary
condition-7BU
4000
Bu all falling on a single line 104 4500
and Derivative (psi)

3800
Pressure Change
Log-Log Match 16BU Simulation 16BU
Pressure (psia)
2 BU 4000
3600 7 BU 103
16 BU 3500
3400 3000
Pressure (psia)
102
2500
3200
10 2000
10-2 10-1 100 101 102 103 1000 1100 1200 1300 1400 1500
3000 Elapsed time (hrs) Elapsed time (hrs)
2800 4 3800
Pressure (psia)
Skin Vs. Rate Horner Match 16BU
3400
2600
3
Skin
3000
2400
2
2600
2200
0 1000 2000 3000 4000 2200
1 100 300 500 700 900 1100 0 1000 2000 3000 4000
Figure 50 Superposition function plots showing reservoir

as non-depleting Figure 54 Analysis Match on Log-Log Pressure Change
& Derivative; Pressure History Simulation; Horner and
Skin vs Rate using Open Ended Rectangle boundary
100
Deconvolved Response
condition-16BU
10-1
Deconvolved Derivative
10-2
Local Grid Refinement
10-3
10-4
Faults: d1=5165ft, d2=2712, d3=1190ft
#(1-16)[16]{1.54660E+05}4076.00 16BU
10-5
10-3 10-2 10-1 100 101 102 103 104
Elapsed Time (hrs)

Figure 51 Comparison of actual normalised pressure 25,000ft
derivative with best deconvolved derivative indicating a

channel behavior Figure 55 Grid model for simulation of well test DST-
Well-15
104 4500
Simulation - Flow Period 2 1
Pressure (psia)
Log-Log Match 2BU 4000

103 0.9
3500
102
3000 0.8
2500 kro
0.7 krg
10 2000
10-2 10-1 100 101 102 103 1000 1100 1200 1300 1400 1500
0.6
K ro /K rg

0.5
2.0 3800
Horner Match 2BU 0.4
Pressure (psia)
Skin Vs. Rate 3600

1.5
3400 0.3
1.0 3200
Skin
0.5 3000 0.2

2800
0.0 2600 0.1
2400
-0.5
100 300 500 700 900 1100 400 800 1200 1600 2000 2400 2800 0
0 0.2 0.4 0.6 0.8
Figure 52 Analysis Match on Log-Log Pressure Change Sg
& Derivative; Pressure History Simulation; Horner and Figure 56 Relative permeability curve used for
Skin vs Rate using Open Ended Rectangle boundary compositional simulation
condition-2BU
18 SPE 116239
4000 Pressure History Comparison
Field test Data 105

Log Log Comparison 7BU

3000
104
Pressure (psia)
Simulated data
2000 103
Simulated Data
(with depth correction) Field test data
102
1000
101 Deviation in derivative match
Pressure dropping to zero

0
0 10 20 30 40 50 60 70 100
10-2 10-1 100 101 102 103
Elapsed time (days)
Elapsed time (hrs)
Figure 57 Comparison of simulated and actual pressure Figure 61 Log-Log Pressure Change and Derivative Plot
history with depth correction Comparison between Simulation and Test Data -16BU
4000
Pressure History Comparison
1
0.9
3000
Simulated Data (with 0.8
depth correction Nc and )
Pressure (psia)
0.7 2BU,, 7BU, 16BU Initial reservoir saturation
O il S a tura tion
2000 0.6
Field test Data
0.5
0.4
1000
0.3 Two-phase region away
1DD from well at end of BU
0.2 260 ft
0 0.1
0 10 20 30 40 50 60 70
0
Elapsed time (days)
10-1 100 101 102 103 104 105
Figure 58 Comparison of simulated and actual pressure Radial Distance from Wellbore (ft)
history with Nc, non-Darcy effect Figure 62 Saturation profile in the region around DST-
Well-15
105
104
Simulated data
103
Field test data
102
100
10-2 10-1 100 101 102 103
Elapsed time (hrs)
Figure 59 Log-Log Pressure Change and Derivative Plot

Comparison between Simulation and Test Data -2BU
105
104
Simulated data
103
Field test data

102
100
10-2 10-1 100 101 102 103
Elapsed time (hrs)
Figure 60 Log-Log Pressure Change and Derivative Plot

Comparison between Simulation and Test Data -7BU

Well Test Analysis in Volatile Oil Reservoirs

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Well Test Analysis in Volatile Oil Reservoirs

Încărcat de

Drepturi de autor:

Formate disponibile

SPE 116239

Well Test Analysis in Volatile Oil Reservoirs

Copyright 2008, Society of Petroleum Engineers

Introduction and background

Multiphase well test analysis

Simulation model for well tests in volatile oil reservoirs

Fluid characterisation and modeling

Relative permeability modeling

Well test behaviors in volatile oil reservoirs

5 DD: constant production rate of 2700 bopd for 5 days

Phase behavior and dynamics

Well test behavior of volatile oil reservoirs

Effect of production rate

Effect of fluid composition

Effect of relative permeability

Effect of capillary number and non-Darcy flow

Well test analysis of volatile oil reservoir

Characterization of volatile oil reservoirs using well test analysis

Interpretation of well test data

Compositional simulation of DST-Well-15

Relative permeability model

Velocity dependence of relative permeability

0.005 cm-1 (8)

Greek : flow potential

Table 3 Composition of fluid sample A

Residual oil 1.25

Residual oil 1.50

CCE Liquid Viscosity

Liquid Viscosity (CP)

K=10mD =15% 100ft

DL1 Liquid Density

DL1 Gas Gravity

1000 2000 3000 4000 1000 2000 3000 4000

Pressure (psig) Pressure (psig)

0.8 Model 1: Kromax=0.95, Krgmax=0.7 =2.5 0.170

Oil Viscosity (cp)

6000 6000 6.0

Oil Relative Mobility (cp-1)

psia 4000 5.8

Elapsed time (days) Radial Distance from Wellbore (ft)

0.7 End of 5DD

Gas Relative Mobility (cp-1)

S<Soi S=Soi S=Soi

D e riva tive (psia )

0.16 102 102

0.14 Volumetric expansion

0.12 5DD 6BU

0.12 Decrease in o due to volumetric expansion

run A-K1-N0-D0 Figure 23 Log-Log pressure derivative showing

Higher Mobility Lower Mobility

Original mobility around well

Oil Viscosity (cp)

103 103 0.15

Mobility around well Increasing viscosity around wellbore

Radial Distance from Wellbore (ft)

Change and Derivative (psia)

Rate Normalised Pressure

Cumulative production of oil =11500 STB for each flow period

Rate Normalised Pressure

3 5DD with Qo =2700 STB/D

Radial Distance from Wellbore (ft)

Radial Distance from Wellbore (ft) Elapsed time (days)

Change and Derivative (psia)

Rate Normalised Pressure