spe 102488 (cox) errors intro multiphase flow cor production anal
TRANSCRIPT
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
1/11
Copyright 2006, Society of Petroleum Engineers
This paper was prepared for presentation at the 2006 SPE Annual Technical Conference andExhibition held in San Antonio, Texas, U.S.A., 2427 September 2006.
This paper was selected for presentation by an SPE Program Committee following review ofinformation contained in an abstract submitted by the author(s). Contents of the paper, aspresented, have not been reviewed by the Society of Petroleum Engineers and are subject tocorrection by the author(s). The material, as presented, does not necessarily reflect anyposition of the Society of Petroleum Engineers, its officers, or members. Papers presented atSPE meetings are subject to publication review by Editorial Committees of the Society ofPetroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paperfor commercial purposes without the written consent of the Society of Petroleum Engineers isprohibited. Permission to reproduce in print is restricted to an abstract of not more than300 words; illustrations may not be copied. The abstract must contain conspicuousacknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract
First and foremost, production analysis techniques requireaccurate rate and bottomhole pressure histories. In most cases
the pressure history of the well is not measured directly at the
bottomhole condition, but is calculated from surface
measurements by the use of single or multiphase flowcorrelations. In some cases significant error is introduced
through the use of these correlations.
This paper evaluates the magnitude of such errors for oil and
gas producers with regard to the estimation of flow capacity,completion efficiency, and effective drainage area. Synthetic
cases are used as control sets in order to evaluate the
sensitivity of the results to the various multiphase flowcorrelations and flowing conditions. In addition to synthetic
(simulated) performance behavior, field cases are presented
and the variance in estimated reservoir and completion
properties is evaluated.
The technical contributions of this paper are:
a. Systematic evaluation of the effect of errors in flow
rates and bottomhole flowing pressures on productiondata analysis using both synthetic and field derived
well performance data.
b. Qualitative guidelines as the effect of errors in rate and
pressure on estimated reservoir properties.
In most cases the flowing bottomhole pressure is not measured
directly, but is calculated based on the measured (surface)tubing pressure profile and flow rate data. Pressure drop
correlations for multiphase flow are used to "convert" thesurface pressure profile into a bottomhole pressure profile
suitable for analysis and interpretation. Depending on the
(multiphase flow) pressure calculation method that is selected,
as well as the producing conditions, significant errors in thecalculated flowing bottomhole pressure can (and will) result
which affects the estimation of reservoir/well parameters.
Three synthetic (simulation) cases were constructed for thepurpose of serving as control data sets for this study. Eachcase represents a single well reservoir producing from the
center of a radial flow geometry system. These simple
scenarios are used to avoid any multiwell (interference)effects, as well as partial or irregular boundaries, etc.
Specifically, the cases considered for this study include:
Fracture stimulated well, low permeability gas reservoir,
Unfractured well, low permeability oil reservoir, andUnfractured well, high permeability oil reservoir.
An appropriate multiphase correlation for each case was
selected to control the simulator. The actual rates and
bottomhole pressures were used to validate the productionanalysis techniques. New bottomhole pressures were then
calculated utilizing different multiphase flow correlations, and
the evaluation process was repeated. The results of eachanalysis were compared to the parameters used in the
simulator in order to establish the conclusions in this paper.As a comprehensive statement, the results of this workhighlight the need for to acquire flowing gradients periodically
to calibrate the multiphase flow correlation.
Background Theory
The use of well production data to characterize reservoirperformance has been utilized by the oil industry for many
SPE 102488
Errors Introduced by Multiphase Flow Correlations on Production AnalysisS.A. Cox and R.P. Sutton, Marathon Oil Co., and T.A. Blasingame, Texas A&M U.
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
2/11
2 SPE 102488
change continuously however, the analytical solutions
typically employed in production analysis require either
constant rate or constant pressure production behavior. In
order to overcome this issue, an equivalent time function mustbe incorporated into the analysis to account for the continuous
changes in rates and pressures. Palacio and Blasingame2
showed that the "equivalent time" function for oil is given bythe following equation:
q
Nt
pe= .........................................................................(1)
For gas wells the "equivalent time" function is defined as:
[ ])(2)(
)()()(
)()()(
0
pmpGz
tqct
pcptdtq
tqct
i
iiig
g
iga = =
........(2)
For the gas well case, the average reservoir pressure profile
must be known (or estimated) in order to calculate the correctequivalent time function. However the drainage area must be
known in order to calculate the average reservoir pressure
from material balance. Therefore, gas production analysis is
inherently iterative in nature.
Various plots are available to determine the effective drainagearea (or volume) of the well. For oil wells, a plot ofdimensionless rate (qDd) versus dimensionless cumulative
production (QDd) will form a straight line with an intercept of
1. For this plot qDdand QDdare defined as:
owa
e
wfi
ooDd q
r
r
ppkh
Bq
4
3ln
)(
1
10x08.7
1
3
=
.........................................................................................(3)
pwfit
oDd N
pphAc
BQ)(
1615.5
=
..............................(4)
For gas wells, a plot of reciprocal dimensionless pressure
(1/pwD) versus dimensionless cumulative production based ondrainage area (QDA) will result in a straight line with an
intercept of 1/2during boundary dominated flow.3 pwDandQDAare defined as:
])()([)(
1
1422
1
wfiwD pmpmtqT
kh
p =
........................(5)
=
)()([
)()([5.4
wfi
i
i
iiDA
pmpm
pmpm
hAp
GzTQ
..........................(6)
During the transient flow period the drawdown history can beused to estimate reservoir flow capacity and completion
this study follows below.
The accuracy of any bottomhole pressure calculation is
dependent on the quality of the input data. The largest
component of pressure drop in a well is normally thehydrostatic pressure constituent, which is dependent on oil,
gas, and water specific gravities, as well as the ratios of eachfluid phase relative to the other phases (GOR, WOR, GLR,
etc).
For multiphase flow, the pressure drop correlations develop
their character from the manner in which the fluid densitygradient is determined. The typical results from a number of
standard correlations for oil and gas producers are shown in
Figs. 1 and 2. These graphs depict the results of 16 flowcorrelations (refs. 4-22) which have been published in theliterature. For reference, Brill23provides further discussion of
these methods, their accuracy and applicability.
For the purpose of this paper, the Hagedorn and Brown16
correlation was used as the basis for simulation of the oil
cases. This method was chosen because the Hagedorn and
Brown method typically predicts pressure gradients that fall
near the midrange value of the solutions offered by all of the
pressure correlations. In other words, the Hagedorn andBrown method provides a consistent response of better than
average accuracy. Using the Hagedorn and Brown methodallows the effect of predicted pressures greater and less than
the actual pressure to be evaluated with production analysis.
For the gas case, the method proposed by Reinicke, et al.21
was used. This method was developed for gas wellsproducing free water.
Simulation Cases (Synthetic Performance)
Three simulation cases were constructed and used as control
sets to test the sensitivity of the production analysis results dueto the errors introduced in the bottomhole pressure estimates
from different multiphase flow correlations. In all cases the
well is in the center of the simulation grid. The gas case uses
a square simulation grid this case includes a hydraulicfracture. The oil cases (unfractured wells) are modeled using
a radial grid. Table 1 presents the reservoir parameters
common to all cases and Table 2 presents the specific
parameters for each case.
Simulation Case 1
Simulation case 1 is the case of a hydraulically-fracturedwell
in a low permeability gas reservoir. The simulation modelwas initialized with a flow capacity of 2 md-ft and an in-place
gas volume of 1,868 MMscf. For this case, the gas production
rate was controlled by setting a maximum rate of 5 000
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
3/11
SPE 102488 3
Table 1 Model Parameters
Parameter Value Unit
Formation top, 10,000 ft
Initial reservoir pressure 5,000 psiaAverage porosity 25 percent
Net pay 40 ftAverage water saturation 30 percent
Reservoir temperature 225 FWater compressibility 3x10-6 1/psiRock compressibility 4.6 x10-6 1/psiWater specific gravity 1.05 g/cc
Oil case PVT properties:Oil gravity 32.7 API
Solution GOR 350 Scf/BblGas specific gravity 0.65 (air = 1)Bubblepoint pressure 2296 psia
Gas case PVT properties:
Gas specific gravity 0.65 (air = 1)
Table 2 Case-Specific Parameters
Case Value Unit
Case 1 - Effective Permeability 0.05 mdCase 1 - Fracture Half Length 200 ft
Case 1 - Fracture Conductivity 500 md-ftCase 2 - Effective Permeability 0.05 mdCase 3 - Effective Permeability 5 md
The "rate-cumulative production" plotting technique was usedto estimate the effective drainage volume for a given well.
This technique was found to be very accurate as the resulting
calculated volume was within 1 percent of the actual volume.
The reservoir flow capacity was estimated to be 1.83 md-ft(using type-curve matching), where this result is within 10
percent of the actual value. Figures 3 and 4present the "typecurve" matches for this case.
The flow rates and tubing pressures obtained from the
simulation model were used to calculate the flowing
bottomhole pressure responses for the well based on industry
accepted multiphase flow correlations. Sixteen different
correlations were used to estimate the flowing bottomholepressure for the well. The new (bottomhole) pressure
estimates were then used to estimate the flow capacity and
effective drainage volume for the well.
The production analysis results for each correlation are
summarized in Table 3. Most of the correlations yielded
acceptable errors in the estimate of flow capacity and in-place
volumes. For the purposes of this work, an unacceptable erroris considered to be an error in excess of 10 percent.
C l ti lti i t th 10 t i i
In general, if the multiphase flow correlation overestimates the
bottomhole pressure, then the estimated flow capacity will
also be high. The slope of the flowing bottomhole pressure
with time during boundary dominated flow controls the in-place volume estimate. An example of this would be the
Kaya17correlation in Fig. 6. This correlation under predicts
flowing pressures, which results in a large negative error inflow capacity, but this approach accurately predicts effective
drainage volume.
Table 3 Case 1: Tight Gas, Low Water Yield
Correlation
kh
(md-ft)
Error
(percent)
OGIP
(Bscf)
Error
(percent)
Simulation 1.83 -8.5 1.86 -0.4Ansari 1.89 -5.5 1.86 -0.4Aziz 1.89 -5.5 1.86 -0.4Baxendell 2.61 30.5 5.66 203.0
Beggs 1.87 -6.5 1.85 -1.0Chierici 2.17 8.5 2.09 11.9Cornish 1.88 -6.0 1.86 -0.4Duns 1.67 -16.5 1.86 -0.4
Fancher 1.94 -3.0 1.89 1.2Gray 1.89 -5.5 1.86 -0.4Griffith 2.20 10.0 2.04 9.2Hagedorn 1.67 -16.5 1.86 -0.4
Kaya 1.45 -27.5 1.86 -0.4Mukherjee 1.89 -5.5 1.86 -0.4Orkiszewski 2.04 2.0 1.96 4.9Poettmann 2.50 25.0 5.66 203.0
Reinicke 1.89 -5.5 1.86 -0.4
In Case 1 the impact of water yield was also explored to
test sensitivity we increased the water yield to 50 Bbl/MMscf.
Despite the increase in water, the calculated bottomhole
pressures were similar to the previous evaluation as illustratedby Fig. 7.
Simulation Case 2
Simulation case 2 is the case of an unfracturedwell in a low
permeability oil case. The model was initialized with a flow
capacity of 200 md-ft and for this case the oil production was
controlled by setting a maximum oil rate of 500 Bopd and aminimum flowing tubing pressure of 50 psia. As noted
earlier, the Hagedorn and Brown16multiphase flow correlation
was used to construct the tubing performance curves for thissimulation run.
The drainage area was estimated from a plot of dimensionless
rate versus dimensionless cumulative production as described
earlier. Figure 8illustrates the typical shape of the qDdversusQDdplot and we note that the effective drainage area for the
well can only be determined using production analysis after
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
4/11
4 SPE 102488
time data should not be used to estimate the drainage area for
this well. We also note that, for this analysis, we assumed
that the total system compressibility was a constant based on
the initial pressure condition. This simplifying assumptionintroduced a slight error in the estimated drainage volume of
the well the effective drainage volume based on the
simulated bottomhole pressure was found to be 6.96 MMblsversus an actual value of 7.4 MMbls (i.e., the input volume)
The flow capacity of the well was estimated from type curve
matching the data in the transient flow period, again using acommercially available production analysis program. The
reservoir flow capacity was found to be 208 md-ft or
approximately 4 percent higher than the input value.The flow rates and tubing pressures from the simulation modelwere used to calculate flowing bottomhole pressure for this
well, utilizing industry accepted multiphase flow correlations.
As in Case 1, sixteen different multiphase correlations wereconsidered in this work. The production analysis results for
each correlation are summarized in Table 4.
Table 4 Case 2: Low Permeability Oil Reservoir
Correlationkh
(md-ft)
Error
(percent)
OOIP
(MMstb)
Error
(percent)
Simulation 208 4.0 6.96 -6.3
Ansari 207 3.5 7.89 6.2Aziz 226 13.0 10.45 40.6Baxendell 225 12.5 7.66 3.1
Beggs 232 16.0 8.12 9.3Chierici 262 31.0 9.75 31.2Cornish 158 -21.0 5.11 -31.3Duns 232 16.0 9.05 21.8Fancher 204 2.0 6.27 -15.6
Gray 166 -17.0 6.04 -18.8Griffith 284 42.0 9.05 21.8
Hagedorn 208 4.0 6.73 -9.4Kaya 220 10.0 8.36 12.5Mukherjee 249 24.5 9.05 21.8Orkiszewski 179 -10.5 7.89 6.2
Poettmann 225 12.5 7.89 6.2Reinicke 208 -17.0 6.96 -18.8
Most of the correlations resulted in unacceptable errors (> 10
percent) in the estimate of flow capacity and in-place volumes.
Only the algorithms of Ansari and Fancher provided estimatesof flow capacity that were within 10 percent error. Five
correlations resulted in in-place fluid volume estimates which
were within 10 percent of the input value.
A performance review of the multiphase flow correlationsprovides some insight into these conclusions. The correlations
higher estimated in-place volume. Conversely, the Cornish10
correlation exhibits a steeper slope, which results in a lower
estimate for in-place volume. Figure 10 also illustrates the
previous conclusion for flow capacity the Chierici, et al.correlation over predicts bottomhole pressure and also over
predicts flow capacity, while the opposite is observed for the
Cornish correlation.
Simulation Case 3
Simulation case 3 represents the higher permeability oil case.
The model was initialized with a flow capacity of 2,000 md-ft.For this case the oil production was controlled by setting a
maximum oil rate of 1,500 Bopd and a minimum flowing
tubing pressure of 50 psia. The Hagedorn and Brown16
multiphase flow correlation was again used to construct the
tubing performance curves for the simulation run.
The drainage area was estimated from a plot of dimensionless
rate versus dimensionless cumulative production (i.e., qDdversus QDd). The effective drainage area for the well can be
determined from production analysis after the well has
achieved boundary dominated flow. Boundary dominated
flow was reached in this case after approximately 5 days of
production (a product of the higher formation permeability)Since boundary dominated flow was reached in such a short
time period, the pressures in the model remained relatively
high and well above the bubblepoint pressure. Fig. 11showsthe bottomhole pressure versus rate performance for the16
multiphase flow correlations evaluated. In this case the
results were in better agreement than the results observed inCase 2 because the primary flow regime in the tubing was
single phase liquid.
It should be noted that the flowing bottomhole pressuredropped below the bubblepoint pressure near the end of the
simulation run, which as with the previous case, caused a gasaccumulation in the near well region. This accumulation
causes a change in the late time slope of the dimensionless rate
versus dimensionless cumulative production plot whichcomplicates the analysis of these data, as illustrated in Fig. 12.
For this analysis we assumed that the total system
compressibility was constant, evaluated at the initial pressure
condition. As with Case 2, this assumption caused a slighterror in the estimated drainage volume of the well. The
effective drainage volume based on the simulated bottomhole
pressure was found to be 6.96 MMbls versus an input value of
7.4 MMbls.
The flow capacity of the well was estimated from type curve
matching the data in the transient flow period using a
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
5/11
SPE 102488 5
Half of the correlations resulted in unacceptable error in the
estimate of in-place volume.
Table 5 Case 3: High Permeability Oil Reservoir
Correlationkh
(md-ft)
Error
(percent)
OOIP
(MMstb)
Error
(percent)
Simulation 1990 -0.5 6.96 -6.3
Ansari 2140 7.0 6.73 -9.4Aziz 2460 23.0 6.96 -6.3Baxendell 1610 -19.5 6.04 -18.8Beggs 2060 3.0 7.20 -3.2
Chierici 2090 4.5 7.66 3.1Cornish 1800 -10.0 5.57 -25.0Duns 1790 -10.5 6.96 -6.3
Fancher 1870 -6.5 6.27 -15.6Gray 1650 -17.5 5.57 -25.0Griffith 2370 18.5 8.36 12.5Hagedorn 1880 -6.0 6.96 -6.3Kaya 1790 -10.5 6.73 -9.4
Mukherjee 1780 -11.0 6.96 -6.3Orkiszewski 2040 2.0 5.80 -21.9Poettmann 1910 -4.5 5.80 -21.9Reinicke 1780 -11.0 5.57 -25.0
Field Example 1
This well is producing from a sandstone reservoir at a depth of15,000 ft with an average porosity of 9 percent, water
saturation of 35 percent, and net pay of 100 ft. The initial
reservoir pressure was 5000 psia. The well was fracture
stimulated upon initial completion, and had an initialproduction rate of approximately 3.0 MMscf/D of wet gas and
120 Bw/D. Typical permeability values for this reservoir
range from 0.01 md to 0.1 md.
No measured bottomhole pressures exist for this well. Two-phase flow correlations were used to estimate the flowing
bottomhole pressure profile for this case and, due to the
condensate and water production from the well, the pressure
correlations resulted in a large variations in the estimatedbottomhole pressures. These results are shown in Fig. 13.
The effective gas permeability obtained using productionanalysis for this case ranged from a low of 0.03 md to 0.2 md
where these values lie with the generally expected range of
permeabilities for this reservoir. The effective fracture halflength ranged from a low of 18 ft for the high permeability
interpretation to a high of 289 ft for the low permeabilityinterpretation. The effective drainage volume obtained from
the normalized pressure plot averaged 1.5 Bscf over all of the
correlations. The results for each correlation are summarizedin Table 6.
Table 6 Field Example 1: Tight Gas
Correlation
kh
(md-ft)
OGIP
(Bscf)
Aziz 4.66 0.64Baxendell 5.16 2.26Beggs 2.70 0.88Chierici 3.97 2.2
Cornish 3.86 1.27Duns 8.16 2.47Fancher 3.00 1.07Gray 5.12 0.94
Griffith 9.15 2.73Hagedorn 4.52 0.89
Mukherjee 5.30 1.26Orkiszewski 18.90 1.49Poettmann 4.38 1.7Reinicke 4.13 1.33
Multiphase flow correlations were used to estimate the
flowing bottomhole pressure profile for the well. Figure 14
illustrates the range of flowing bottomhole pressures
calculated for the well. The effective gas permeability
obtained from production analysis for this well ranged from a
low of 4.5 md to 7.1 md, with and average of 5.6 md. Theeffective drainage volume obtained from the normalized
pressure plot ranged from 1.34 MMbo to 2.91 MMbo and
averaged 1.5 Bscf considering all correlations. The results foreach correlation are summarized in Table 7.
Table 7 Field Example 2: Low Permeability Oil
Correlation
kh
(md-ft)
OOIP
(MMstb)
Aziz 194 2.91
Baxendell 186 2.50Beggs 186 2.91Chierici 222 2.77
Cornish 156 1.38Duns 203 2.29Fancher 203 1.59Gray 163 1.47
Griffith 243 2.68Hagedorn 156 1.57Mukherjee 243 2.77Orkiszewski 163 1.63
Poettmann 186 2.51Reinicke 163 1.58
Observations and Conclusions
1. Multiphase flow correlations that over predict flowingbottomhole pressure also over predict flow capacity. The
converse is true for methods that under predictb h l
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
6/11
6 SPE 102488
relationship remains important for the correct
determination of in-place volume.
4. The limitations of multiphase flow correlations need to be
recognized so that these correlations can be used properly.Accurate production rates and pressures must be
recorded, as errors in the rates and pressures affect the
material balance of the system, as well as estimates ofbottomhole pressure obtained from correlations. Periodic
flowing pressure surveys are recommended to ensure that
the pressure correlations are properly calibrated. In theabsence of (measured) pressure surveys, the authors offer
the following recommendations for computing/estimating
flowing bottomhole pressures using multiphase flow
correlations:
Gas Wells
Reinicke, Remer, HueniGrayHagedorn and BrownCornish
Oil Wells
Hagedorn and Brown
Duns and RosBeggs and BrillOrkiszewskiKaya
These recommendations are offered only as a guide we
believe that it is helpful to examine the results using
several methods in order to determine a range ofuncertainty.
Acknowledgments
We thank the management of Marathon Oil Company for
permission to print this article. Acknowledgment is due tovarious colleagues for providing production data for the field
cases.
Nomenclature
A = drainage area, ftBo = oil formation volume factor, RB/STB
ct = total compressibility, psi-1
G = original gas in place, Mscfh = reservoir thickness, ft
k = effective permeability to gas, md
m p( ) = real gas pseudo pressure, psi2/cp
m p( ) = m p m pi( ) ( ) , psi2/cp
Np = cumulative oil production, bbl
pwf = bottomhole producing pressure, psia
te = equivalent time oil, days
zi = gas compressibility factor atpi
= porosity, fraction
= viscosity, cp = 3.14159Subscripts
i = initial
g = gas
References
1. Arps, J.J. : "Analysis of Decline Curves", Trans., AIME(1945) 160, 228-47.
2. Palacio, J.C. and Blasingame, T.A.: "Decline-CurveAnalysis Using Type Curves Analysis of Gas Well
Performance Data," paper SPE 25909 presented at the
1993 Rocky Mountain Regional Meeting/LowPermeability Reservoirs Symposium and Exhibition,
Denver, 26-28 April.
3. Agarwal, R.G., Gardner, D.C., Kleinsteiber, S.W. andFussell, D.D.: "Analyzing Well Production Data Using
Combined-Type-Curve and Decline-Curve AnalysisConcepts," SPEREE(October 1999) 478.
4. Ansari, A.M., Sylvester, N.D., Shoham, O., and Brill, J.P.:"A Comprehensive Mechanistic Model for Upward Two-
Phase Flow in Wellbores," paper SPE 20630 presented atthe 65thAnnual Technical Conference and Exhibition, New
Orleans, LA (Sept. 23-26, 1990).
5. Aziz, K.: Ways to Calculate Gas Flow and Static Head,Handbook Reprint from Pet. Eng., Dallas, TX (1963).
6. Aziz, K., Govier, G.W., and Fogarasi, M.: "Pressure Dropin Wells Producing Oil and Gas," J. Cdn. Pet. Tech.
(July-Sept., 1972) 38-48.7. Baxendell, P.B. and Thomas, R.: "The Calculation ofPressure Gradients in High-Rate Flowing Wells," J. Pet.
Tech.(Oct., 1961) 1023-1028.
8. Brill, J.P. and Mukherjee, H.: Multiphase Flow in Wells,Monograph 17, SPE, Richardson, TX (1999).
9. Chierici, G.L., Ciucci, G.M. and Sclocchi, M.:"Two-Phase Vertical Flow in Oil Wells - Prediction ofPressure Drop,"J. Pet. Tech.(Aug., 1974) 927-938.
10.Cornish, R.E.: "The Vertical Multiphase Flow of Oil andGas at High Rates,"J. Pet. Tech. (July, 1976) 825-831.
11.Cullender, M.H. and Smith, R.V.: "Practical Solution ofGas-Flow Equations for Wells and Pipelines with Large
Temperature Gradients," Trans. AIME (1956) Vol. 207,281-287.
12.Duns, H., Jr. and Ros, N.C.J.: "Vertical Flow of Gas andLiquid Mixtures in Wells " Proc 6th World Petroleum
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
7/11
SPE 102488 7
Two-Phase Flow in Small-Diameter Vertical Conduits," J.
Pet. Tech.(April, 1965) 475-484.
17.Kaya, A.S., Sarica, C., and Brill, J.P.: "Mechanistic
Modeling of Two-Phase Flow in Deviated Wells," SPEProduction and Facilities, (Aug., 2001) 156-165.
18.Mukherjee, H. and Brill, J.P.: "Liquid Holdup Correlationsfor Inclined Two-Phase Flow," J. Pet. Tech. (May, 1983)1003-1008.
19.Orkiszewski, J.: "Predicting Two-Phase Pressure Drops inVertical Pipe,"J. Pet. Tech.(June, 1967) 829-838.
20.Poettmann, F.H. and Carpenter, P.G.: "The MultiphaseFlow of Gas, Oil, and Water Through Vertical Flow
Strings with Application to the Design of Gas Lift
Installations,"Drill. and Prod. Prac.,API (1952) 257-317.21.Reinicke, K.M., Remer, R.J., and Hueni, G.: "Comparison
of Measured and Predicted Pressure Drops in Tubing for
High-Water-Cut Gas Wells," paper SPE 13279 presented at
the 59th Annual Technical Conference and Exhibition,
Houston, TX (Sept. 16-19, 1984).22.Ros, N.C.J.: "Simultaneous Flow of Gas and Liquid as
Encountered in Well Tubing," J. Pet. Tech. (Oct., 1961)1037-1049.
23.Beggs, H.D. and Brill, J.P.: "A Study of Two-Phase Flowin Inclined Pipes,"J. Pet. Tech.(May, 1973) 607-617.
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
8/11
8 SPE 102488
Fig. 1 Multiphase Flow Correlations Oil
Fig. 4 Sim. Case 1 Rate Cumulative Decline Curve
Effective Drainage Area 40 Acres
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
QDA
1/PwD
Actual
Analytical
Fig. 3 Sim. Case 1 - Type-curve match
0.1
1
10
100
1000
0.0001 0.001 0.01 0.1 1 10 100
tDA
PwD
orPwD'
Actual PwD
Actual PwD'
Analytical PwD
Analytical PwD'
Infinite Conductivity Fracture in 1 to 1 Rectangular Boundary at 1 year
Match Simulation
Kh = 1.83 md-ft, 2.0 md-ft
Xf = 209 ft, 200 ft
Area= 40 Acres, 40 Acres
Fig. 2 Multiphase flow correlations gas
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
9/11
SPE 102488 9
Fig. 7 Sim. Case 1 - Water rate sensitivity comparison
Fig. 6 Sim. Case 1 - Multiphase flow Correlation Comparison
0
1000
2000
3000
4000
0 50 100 150 200 250 300 350 400
Time, Days
BottomholePressure,
Psia
Simulation
KAYA
Fig. 8 Sim. Case 2 - Normalized rate/cumulative production
0
0.5
1
1.5
2
0 0.25 0.5 0.75 1
QDd
qDd
Time to BDF Match Simulation
Volume 6.96, 7.4 MMbo
0
1000
2000
3000
4000
0 50 100 150 200 250 300 350 400
Time, Days
BottomholePressure,
Psia
Simulation
Baxendell & Poettman
Chierici
Fig. 5 Sim. Case 1 Multiphase Flow Correlation Comparison
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
10/11
10 SPE 102488
Fig. 10 Sim. Case 2 - Correlation comparison late time
Fig. 11 Sim. Case 3 Correlation comparison Fig. 12 Sim. Case 3 - Normalized rate vs. comumulative
production
0
0.5
1
1.5
2
0 0.25 0.5 0.75 1
QDd
qDd
Time to BDF Match Simulation
Volume 6.96, 7.4 MMbo
Fig. 9 Sim. Case 2 - Correlation comparison early time
-
7/25/2019 SPE 102488 (Cox) Errors Intro Multiphase Flow Cor Production Anal
11/11
SPE 102488 11
Fig. 14 Field Case 2 Calculated Bottomhole Pressures
0.0
1000.0
2000.0
3000.0
4000.0
5000.0
0 10 20 30 40 50 60 70
Time, Days
BottomholePressure,
Psia
CHIERICI
CORNISH
FANCHER
Fig. 13 Field Case 1 -- Calculated Bottomhole Pressures
0.0
1000.0
2000.0
3000.0
4000.0
5000.0
6000.0
7000.0
0 20 40 60 80 100 120 140
Time, Days
B
ottomholePressure,
Psia
AZIZ
ORKISZEWSKI
REINICKE