E Mplete Issue Ore Information About This Article Urnal's Homepage in
E Mplete Issue Ore Information About This Article Urnal's Homepage in
E Mplete Issue Ore Information About This Article Urnal's Homepage in
mplete issue
urnal's homepage in
dalyc.org
CT&F C
How to
ISSN: 0122-5383
[email protected]
ECOPETROL S.A.
Colombia
1,2
Universidad Surcolombiana, Programa de Ingeniería de Petróleos, Grupo de Investigación en Pruebas de
Pozos, Neiva,
Huila, Colombia
3
Ecopetrol S.A.-Instituto Colombiano del Petróleo, A.A. 4185 Bucaramanga, Santander,
Colombia
e-mail: [email protected] e-mail: [email protected] e-mail: [email protected]
T he
to conditions satisfy
gas flow
gas are
equation is
behavior insignificant, when
normally
In this paper, we implemented the Tiab’s Direct Synthesis (TDS) technique, to include pseudotime
effects, and observe its influence on the interpretation results of gas well test data at early and late
time periods. New analytical equations to estimate reservoir permeability, wellbore storage
coefficient, pseudoskin factor and reservoir drainage area are presented. Then, a comparison of
results against rigorous time was carried out for simulated and field cases. We found acceptable
results for permeability, pseudoskin factor and wellbore storage coefficient. However, for the case of
reservoir drainage area, the deviation error was of 4,1% for a simulated case and 17,9% for a field
case. However, the smaller of these deviations may be small if related to pressure transient analysis
results. However, this deviation in a gas reservoir with reserves of one tera standard cubic feet is
equivalent to a huge difference of 38 gigas of standard cubic feet of gas which may have an
economic impact to any oil company.
Keywords: reservoir gas, pressure, TDS technique, steady state, permeability, radial flow, mathematical
models.
difusividad
de los líquidos satisfaga el comportamiento del gas cuando se analizan pruebas de
presión en yacimientos gasíferos. Las pruebas de declinación de presión se analizan mejor usando
la función pseudopresión, cuando los efectos de almacenamiento de pozo son insignificantes. Por
otra parte, las pruebas de restauración de presión requieren la linealización tanto de la
pseudopresión como del pseudotiempo. Sin embargo, cualquiera que sea el caso, la función de
pseudotiempo presenta ciertos efectos a tiempos de prueba muy largos en formaciones de
permeabilidad moderada a alta.
En este artículo, implementamos la técnica de Síntesis Directa de Tiab, (TDS), para incorporar los efectos del pseudotiempo,
y observar su influencia en los resultados de interpretación de pruebas de presión en yacimientos de
gas a tiempos tempranos y tardíos. Se desarrollaron nuevas ecuaciones analíticas para la
estimación de la permeabilidad del yacimiento, el coeficiente de almacenamiento del pozo, el factor
de pseudodaño y el área de drenaje del pozo. Luego, para casos de campo y simulados, se efectuó
una com- paración de los resultados contra aquellos donde se usa el tiempo riguroso o normal.
Encontramos valores aceptables de permeabilidad, pseudo factor de daño y coeficiente de
almacenamiento. Sin embargo, para el área de drene del pozo, la desviación fue de 4,1 y de 17,9%
para un caso de campo. La menor de estas desviaciones es un número que resulta pequeño si lo
relacionamos con los resultados producidos en la interpretación de pruebas de presión. Sin
embargo, esta desviación en un yacimiento con reservas de un tera de pies cúbicos a condiciones
normales equivale a una enorme diferencia de 38 gigas de pies cúbicos a condiciones normales lo
cual puede impactar económicamente a cualquier empresa.
Palabras clave: yacimientos de gas, presión, técnica TDS, estado estacionario, permeabilidad, flujo radial,
modelos matemáticos.
CT&F - Ciencia, Tecnología y Futuro - Vol. 3 Núm. 3 Dic. 2007 114
NOMENCLATURE
A Drainage area, ft2 B Oil volume factor, bbl/STB (crude), bbl/Mpcn (gas) C
Wellbore storage coefficient, bbl/psi D NonYDarcy flow coefficient, (Mscf/D)-1 c
Compressibility, 1/psi ct Total compressibility, 1/psi h Formation thickness,
ft k Permeability, md m(p) Pseudopressure function, psi2/cp p Pressure, psi
qg Gas
Well radius, ft s Skin factor s, Pseudoskin factor T
flow rate, Mscf/D rw
Temperature, oR t Time, h ta(p) Pseudotime function, psi h/cp W ̈P S ¶
Pseudopressure derivative function with respect to rigorous/real time ta ̈P S ¶
Pseudopressure derivative function with respect to pseudotime Z Gas deviation
factor
GREEK
̈ Change, drop Ø
Porosity, fraction μ
Viscosity, cp
SUBINDEXE
S
FREDDY-HUMBERTO ESCOBAR
et al.
6
CT&F - Ciencia, Tecnología y Futuro - Vol. 3 Núm. 3 Dic. 2007
11
S.I. METRIC CONVERSION FACTORS
(2
EFFECT OF THE PSEUDOTIME FUNCTION ON GAS RESERVOIR
Notice properties. that ȝ In their and analysis ct are
now pressure dependent for the application of the TDS
technique of gas reservoirs, Nuñez et al. (2003) used the following dimensionless quantities:
(3)
(4)
(5)
tion Including 3, the dimensionless the pseudotime pseudotime function, is ta given
(p), in by:
Equa-
(6)
Notice that the viscosity-compressibility product is not seen in Equation 4 since they are included in
the pseudotime function. However, if we multiply and, then, divide dimensionless by time (ȝFt)i expression,
a similar equation to the Equation 3, general will be obtained.
(7)
As presented by Tiab (1993, 1995), the governing equation for the well pressure behavior during radial flow is
expressed by:
(8)
From a log-log plot of pseudopressure and pseu- dopressure derivative against pseudotime, Figure 1,
several main characteristics are outlined;
1. The early unitYslope line originated by wellbore storage is described by the following equation:
(9)
Figure 1. Log-log plot of pseudopressure and pseudopressure derivative vs. pseudotime
Replacing the dimensionless parameters in Equa- tion 9, a new equation to estimate the wellbore storage
coefficient is obtained:
(10)
2. The intersection of the early unitYslope line with the radial horizontal straight line gives:
(11.a)
From the above relationship is obtained an equation to estimate either permeability or wellbore storage
once the dimensionless parameters are replaced:
(11.b)
3. According to Tiab (1993, 1995), another form of Equation 8 is obtained when wellbore storage and skin
factor are included:
(12)
From the above equation, the derivative of pseudo- pressure with respect to ln tDa/CD:
(13)
CT&F - Ciencia, Tecnología y Futuro - Vol. 3 Núm. 3 Dic. 2007 117
FREDDY-HUMBERTO ESCOBAR et al.
FromEquation 4, the dimensionless pseudopressure
118 derivative with respect to ln tDa/CD gives:
(14)
Combination of Equation 12 and Equation 13 will result into an equation to estimate permeability:
(15)
4. Dividing Equation 11.a by Equation 12, replacing the dimensionless quantities in the resulting
expres- sion and, then, solving for the pseudoskin factor will yield:
(16)
The pseudosteady state solution of the diffusivity equation of closed systems is expressed by:
(17.a)
As suggested by Equation 17.a, the analysis in this work is only applicable for circular drainage areas,
then, we have to consider limitations to other reservoir shapes. The pseudopressure derivative of the above
equation is then:
(17.b)
The intersection point of the above straight line and the radial flow line is:
(18)
Alter substituting the dimensionless pseudotime function into Equation 18, a new equation for the well
drainage area is presented:
(19)
For comparative purposes, we cite the following equations developed by Nuñez et al. (2003).
(20)
(21)
(22)
The following equation is taken from Nuñez et al. (2002).
(23)
STEP-BY-STEP PROCEDURE
Detailed explanation of the appropriate use and ap- plication of the TDS technique are presented by
Tiab (1993, 1995), Tiab (1993, 1994), Nuñez et al. (2002), Nuñez et al. (2003). A summarized procedure is
out- lined as follows:
Step 1. Construct a logYlog plot of Am(p) and talAm(p)' versus ta.
Step 2. Identify the different flow regimes and draw the horizontal line along both radial flow and the
unitYslope line. For the later case, the lines is drawn through both the wellbore storage points and the late
pseudosteadyYstate regime, if given the case.
Step 3. Read the intersection point of the wellbore storage line and the radial that: wellbore Am(p)storage i =
(ta coefficient ̈P(p)’)i flow
= from (ta line,
̈P(p)’)Equation (tai ̈P(p)i), r note
and find the 10 and the
permeability from Equation 15. Permeability ought to be verified with Equation 11.b using the pseudotime
of intersection of the early unitYslope and the radial lines.
a ̈P(p)’)r =
5’510,526
.
Pa s)ta(p)r 247 402,7 psi h/cp (6,141l1012 s KPa/
.
tr 2,015
h (7254 s) ̈P(p)r 54Г119 360 psi2/cp (3,520*1012 KPa2/Pa s)
Figure 2a. Log-log plot of pseudopressure and pseudopressure derivative vs. time for the field example 1
Using Equation 15, 10 and 16, permeability of 8,519 md, wellbore storage coefficient of 0,012 Mscf/ psi
(0,04928 m3/KPa) and pseudoskin factor of Y0,71 were respectively estimated. These same parameters, see
Table 2, were also found for the actual time using Equations 20, 21 and 22. Table 2 displays the results of
interpreting the test by means of non-linear regres- sion analysis using a commercial software. We are not
satisfied with these results because of poor matching
Table 1. Reservoir, gas and well properties for the worked examples
(1) Determined by non-linear regression analysis from a commercial software (2) From Spivay and Lee (1986)
FIELD
FIELD EXAMPLE 1
EXAMPLE 2
SIMULATED EXAMPLE 2
Paramater Value
pi , psia 3000 7800 1062 4500 5000 k, md 8,51(1) 0,13(2) - 10 30
S’ -1,42(2) 3,5(2) - 5 -
C, Mscf/psi 0,0247(3) 0,1(2) - 2 -
R 660 705 570 672 672
T, o
qg , M
scf/D 1000 415 10000 5000 5000
h, f t 10 14,5 40 100 50
ø, % 19 12,5 16 5 5
cti, 1/psi 2,35x10-3 6,45x10-4 0,00103 1,18x10-4 1,32x10-4 μi, cp 0,02024 0,0306 0,013 0,03036 0,026647
Ȗg 0,65
, ft 0,365 0,3 0,3 0,3 0,3 A, A
0,58 0,9 0,7 rw c - - - 14128 -
CT&F - Ciencia, Tecnología y Futuro - Vol. 3 Núm. 3 Dic. 2007 119
FIELD EXAMPLE 3
SIMULATED EXAMPLE 1
FREDDY-HUMBERTO ESCOBAR et al.
120
for this test is reported in Table 1. Estimate reservoir permeability, pseudoskin factor, and wellbore storage
coefficient using the TDS technique for both real time and pseudotime.
Solution
From the pseudopressure and pressure derivative plots, Figures 3a and 3b the following data are read:
.
ti 0,162 h (583,2 s) ta (p)i 81 129,7 psi h/cp (2,014l1012 s KPa/Pa s)
.
̈P(p)psi2/cp (5,626*10i = (ta ̈P(p)’)12 KPai 2 = /Pa(ta s)
̈P(p)’)r 118Г338
700
between the actual pressure trend and the simulated solution.
.
Pa s)ta(p)r 50Г269 174,4 psi h/cp (1,248l1015 s KPa/
Field example 2
tr 350,9
h (1,26l106 s)
Spivey and Lee (1986) presented a drawdown test ran in a gas reservoir. The relevant information
.
Pa s)
̈P(p)r 3 670Г356 000 psi2/cp (1,745l1014 KPa2/
Figure 2b. Log-log plot of pseudopressure and pseudopressure derivative vs. pseudotime for the field example 1
Table 2. Results of the worked examples
REAL TIME PSEUDOTIME SIMULATOR
ABS. DEVIATION WITH RESPECT PSEUDOTIME
Parameter Field Example 1
k, md 8,52 8,519 8,51 -0,01
S’ -0,1 -0,71 -1,42 85,48
C, Mscf/psi 0,02 0,012 0,0247 86,85
Field Example 2
k, md 0,12 0,12 0,13 0,00
S’ 8,54 9,17 3,50 6,85
C, Mscf/psi 0,01 0,0054 0,10 -84,36
Field Example 3
k, md 1366 1356 1310 0,73
S’ 1,33 1,39 7,3 4,31
A, Ac 388 329 336 17,93
Simulated Example 1
k, md 12,481 12,481 10 0,00
S’ 5,284 5,311 5 0,51
C, Mscf/psi 2,180 2,180 2 0,01
Simulated Example 2
A, Ac 14670,9 14093,2 14128,4 4,10
CT&F - Ciencia, Tecnología y Futuro - Vol. 3 Núm. 3 Dic. 2007
Using Equation 15, 10 and 16, permeability of 8,52 md, Figure 4a. Log-log plot of pseudopressure and
pseudopressure derivative vs. time for the field
wellbore storage coefficient of 0,0054 Mscf/ psi
example 3
(0,02218 m3/KPa) and pseudoskin factor of Y0,71
were respectively estimated. Same calculations were
performed with using Equation 20 through
Equation 22, as reported in see Table 2. That table
presents not only the results from this study but
also those from Spivay and Lee (1986) which were
found by typeYcurve matching.
Field example
3
A drawdown test was run in a unique well of a Colombian
reservoir with pseudopressure and pseudo- pressure
Figure 4b. Log-log plot of pseudopressure and
derivative reported in Figure 4a. Information pseudopressure derivative vs. pseudotime for the field
concerning well, fluid and reservoir characteristics example 3
Solutio .
psi2/cp (6,506*1010 KPa2/Pa s) (ta ̈P(p)’)r 74
n
2 9 2 .
750 psi
The pseudopressure and pressure derivative plots are given in Figures /cp (3,553*10
4a and 4b for real KPa
time /Pa
ands) ta(p)rpi =
pseudotime,
respectively. The following data are read:
.
510 000 h (1,266*1013 s KPa/Pa s)
Figure 3a. Log-log plot of pseudopressure and
.
tr 1,97 h (7 092 s) Am(p)r 1Г370 865 psi2/cp (6,517l1010 KPa2/Pa s) (W ̈P(p)’)r 74 203 psi2/cp (3,572l10
pseudopressure
9
derivative vs.KPa
2
time for/Pa s)
the .field
example 2
10 .
̈P(p)psi2/cp (9,102l10i = 2
(ta ̈P(p)’) KPa i /Pa=
(t s)
a ̈P(p)’)r 1Г914,726
Pa.s)ta(p)r 343Г851 894 psi h/cp (8,535l1015 s KPa/
6 .
tr 1 300 h (4,68l10 s) ̈P(p)r 58Г922
620 psi2/cp (2,801l1012 KPa2/Pa s)
From Equation 15 a permeability of 12,48 md was determined. It was also estimated a pseudoskin factor of
5,31 with Equation 16 and a wellbore storage coef- ficient of 2,18 Mscf/psi (8,953 m3/KPa) with Equation 10.
Practically the same results were obtained using Equations 28.b, 18 and 28.c from the reference of Nuñez
et al. (2003).
Simulated example 2
A pressure drawdown test in a closed circular gas reservoir was also simulated using a commer- cial software
based upon the information of Table 1. Estimate the area of this reservoir using the TDS technique and
compare to the reference area given in Table 1.
Solution
The simulated values of pseudopressure and pseudopressure derivative against either real time or
pseudotime are presented in Figures 6a and 6b, respectively. The intersection point of the radial and late
.
pseudosteady-state lines is found at a pseudotime, Pata(p) s) priand , of the 308Г763 is found to be Figure
5a Log-log plot of pseudopressure and pseudopressure
derivative vs. time for the synthetic example 1
737,3 h psi/cp (7,664l10 s KPa/ corresponding 1133 hours (4,076l10real intersection 6 s) from Figure time,
18
t4b. pri,
8 2
The area with the rigorous time using Equation 23 was of 14 670,9 Ac. (5,938l10 m ). From Equation 19 the
estimated area was 14 093 Ac. (5,704l108 m2).
ANALYSIS OF RESULTS
It is observed, from the worked examples, a good agreement between the simulated values of permeabil- ity,
pseudoskin factor and wellbore storage obtained
Figure 5b Log-log plot of pseudopressure and pseudopressure derivative vs. pseudotime for the synthetic example 1
CT&F - Ciencia, Tecnología y Futuro - Vol. 3 Núm. 3 Dic. 2007
EFFECT OF THE PSEUDOTIME FUNCTION ON GAS RESERVOIR
using the TDS technique when using either regular time or pseudotime. Practically, same values of
permeability, pseudoskin factor and wellbore storage are found no matters if either rigorous time or
pseudotime is uti- lized. However, as expected and exposed formerly in the literature, the estimation of the
reservoir drainage area has a higher deviation resulting from the use of real time. The resulting drainage
area using real time has an absolute deviation of 4,1% while the one with pseudotime has a deviation of 3,84%
with respect to the simulated value. Also the deviation of the area when employing pseudotime has a deviation
of 0,25% with respect to the simulated value. From field example 3 was found that the drainage area has an
absolute devia- tion of 17,9% which ir more remarkable. This confirms that the influence of pseudotime is more
relevant for long producing times. Although, the deviation value
of 3,84% may fall into a widely accepted range of the daily measurements normally obtained from well test
analysis, this can introduce significant differences in estimation of reserves.
The non-linear regression analysis simulation conducted using a commercial software did not match
perfectly the pseudopressure and pressure derivative curve, therefore, the results may not be as
accurate as expected. Consequently, the field cases were not treated in the analysis. Nonetheless, the results of
the permeability reference values agree quite well with Figure 6a. Log-log plot of pseudopressure and pseudopressure
derivative vs. time for the synthetic example 2
those from the TDS technique.
CONCLUSIONS
Estimation
of the drainage area in gas bearing for-
mations is affected when the time is used rigorously.
We found absolute differences between 4,1 and 17,9% which impacts also gas reserves. Therefore, to assure
accuracy, care should be taken into account when interpreting long drawdown pressure tests. For either
relatively short drawdown tests or highly low permeability gas formations, regular time may be used for
practical purposes without incurring in a significant error.
Additionally,
the TDS technique has been extended
by incorporating new analytical solutions, which in-
clude the pseudotime function, for the estimation of formation permeability, pseudoskin factor, wellbore
storage coefficient and reservoir drainage area. Field and synthetic examples were carried out to test the
sensitivity of the mentioned parameters when the pseudotime function is included, as compared to the
actual time.
ACKNOWLEDGMENTS
The authors gratefully acknowledge the finan- cial support of Ecopetrol S.A. Yinstituto Colom- biano del
Petroleo (ICP), under the mutual agree- ment number 008, signed between this institution and Universidad
Surcolombiana (Neiva, Huila, Colombia).
CT&F - Ciencia, Tecnología y Futuro - Vol. 3 Núm. 3 Dic. 2007 123
Figure 6b. Log-log plot of pseudopressure and pseudopressure derivative vs. pseudotime for the synthetic example 2
FREDDY-HUMBERTO ESCOBAR et al.
REFERENCES
Agarwal, G. (1979). Real gas pseudoYtime a new function for presssure buildup analysis of MHF gas wells. Paper
presented at the SPE 54th Technical Conference and Ex- hibition of the Society of Petroleum Engineers of AIME,
Las Vegas, NV, Sept. 23Y26. SPE 8279.
AlYHussainy, R., Ramey, H.J., & Crawford, P.B. (1966). The flow of real gases through porous media. J. Petroleum
Technol., May. 624Y636, Trans. AiME, 237.
Aminiam, K., Ameri, S., W.E., Abbitt, W. E., & Cunningham, L.E. (1991). Polynomial approximations for gas pseudo-
pressure and pseudotime. SPE Eastern Regional Meeting, Lexington, Kentucky, Oct. 22Y25. SPE 23439.
Graham, J. W., & Warwick, M. K. (1984). Scaled – time: a new pseudotime function for approximately linearizing
gas reservoir models. Unsolicited Paper SPE 13418.
Lee, J., 1982. Well Testing. SPE textbook series (Vol. 1).
Lee, W. J., & Holditch, S. A. (1982). Application of pseu-
dotime to buildup test analysis of lowYpermeability gas wells with longYduration wellbore storage distortion. J.
Petroleum Technol., 2877Y2887.
Nuñez, W., Tiab, D. & Escobar, F. H. (2002, October). Análisis de presiones de pozos gasíferos verticales con
fracturas de conductividad infinita en sistemas cerrados sin emplear curvas tipo. Boletín Estadístico Mensual del
ACIPET, pp.9Y14.
Nuñez, W., Tiab, D., & Escobar, F.H. (2003). Transient pressure analysis for a vertical gas well intersected by a
finiteYconductivity fracture. SPE Production and Op- erations Symposium, Oklahoma City, OK, March 23Y25. SPE
80915.
Spivey, J. P., & Lee, W. J. (1986). The use of pseudotime:
wellbore storage and the middle time region. SPE Un- conventional Gas Technology Symposium, Louisville,
Kentucky , May 18Y21. SPE 15229.
Tiab, D. (1993). Analysis of pressure and pressure deriva-
tive without typeYcurve matching: 1Y skin factor and wellbore storage. SPE Production Operations Sympo- sium,
Oklahoma City, OK. March 21Y23, SPE 25423, 203Y216. Also, J. Petroleum Scien. and Engineer., (1995), 12:
171Y181. 1 24Tiab, D. (1993b) Analysis of pressure derivative without typeYcurve matching: vertically fractured wells in
closed systems. SPE Western Regional Meeting, Anchorage, Alaska, May 26Y28. Also, J. Petroleum Scien. and
Engi- neer., 11 (1994), 323Y333. SPE 26138.
CT&F - Ciencia, Tecnología y Futuro - Vol. 3 Núm. 3 Dic. 2007