Comparison of PVT Correlations For Predicting Crude Oil Properties: The Brazilian Campos Basin Case Study
Comparison of PVT Correlations For Predicting Crude Oil Properties: The Brazilian Campos Basin Case Study
net/publication/336354729
CITATIONS READS
0 102
2 authors:
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Patrick V. Mangili on 09 October 2019.
a
Mangili, P. V. 1; a Ahón, V. R. R.
a
Fluminense Federal University, Department of Chemical and Petroleum Engineering, Niterói - RJ - Brazil
Received: 15.01.2019 / Revised: 10.07.2019 / Accepted: 17.07.2019 / Published on line: 08.10.2019
ABSTRACT
Pressure-Volume-Temperature correlations are essential for estimating the required parameters
necessary for the identification of reservoir fluid properties when experimental laboratory data are not
readily available. Although several models have been published for different oil fields, there is a lack of
studies addressing the Brazilian pre-salt region. Hence, in this paper we compared twenty empirical
correlations for the determination of the solution gas-oil ratio, oil formation-volume-factor, and under-
saturated oil viscosity for Brazilian pre-salt oil samples collected from the Campos Basin region, originally
evaluated by Elias and Trevisan (2016). From our statistical results, some models presented good
estimation performances when compared to said reference’s results. In fact, through a statistical analysis,
Al-Shammasi (2001)’s correlation proved to be the best estimation method for solution gas-oil ratio,
whereas Al-Marhoun (1985) and Beal’s (1946) correlations were deemed the most accurate for the
prediction of oil formation-volume-factor and under-saturated oil viscosity, respectively.
KEYWORDS
Campos Basin; oil formation-volume-factor; PVT correlations; gas-oil ratio solution; under-saturated oil
viscosity
1
To whom all correspondence should be addressed.
Address: Fluminense Federal University, Department of Chemical and Petroleum Engineering, Niterói - RJ - Brazil.
ZIP Code: 24210-240 | e-mail: pmangili@id.uff.br
doi:10.5419/bjpg2019-0013
129
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
In light of such dynamics, the oil and gas 2.1 PVT properties
companies have been committed to either As previously discussed, reservoir engineers
developing new techniques or improving already-
frequently are challenged to design production
existing strategies to produce the so-called “black
facilities to process crude oil so as to achieve the
gold.” Therefore, understanding the phase
highest economic performance. They must,
behavior of hydrocarbons present in the reservoir
therefore, understand the phase behavior of
is of utmost importance, since it is highly necessary
hydrocarbons throughout the entire lifetime of the
for process design (Rasouli et al., 2008), evaluation reservoir and be able to estimate fluid properties
of reservoir performance (Elsharkawy et al., 1995),
that depend on different conditions of
and field development (Torabi et al., 2014). The temperature and pressure. According to Ahmed
thermodynamic behavior of petroleum fluids can
(2010), these properties are determined preferably
be determined through Pressure-Volume- through laboratory analyses that consist in
Temperature (PVT) tests, which are conducted on
studying the volumetric behavior of live oil
laboratory samples collected in situ. PVT analyses
samples, which are, then, characterized by being
are essential for providing the required parameters
subjected to several pressure level changes in a
necessary for the identification of reservoir fluid
PVT cell.
properties such as bubble point pressure (Pb),
solution gas-oil ratio (Rs), and oil formation- The results obtained from experimental PVT
volume-factor (Bo) (Karimnezhad et al., 2014; tests are essential for geophysics and engineers to
Salehinia et al., 2016). Nevertheless, PVT analyses perform material balance calculations and
depend on experimental data that may not be reservoir simulation studies, identify potential
readily available. In this regard, empirical risks, determine reserve recovery, and develop
correlations may be conveniently useful for production plans (Mansour et al., 2013; Potsch et
estimating such parameters, which can be carried al., 2017). For this reason, oil samples must
out by using field measured data such as preserve the same characteristics as their original
temperature, oil gravity, and gas density, among formation to provide reliable, high quality data,
others (Gharbi & Elsharkawy, 2003). and, thus, “reduce uncertainty” and “improve
investment efficiency” (Nnabuo et al., 2014).
This paper aims to demonstrate the significance
of applying PVT correlations in the estimation of However, as mentioned earlier, experimental
reservoir fluid properties in the absence of data may not be available, especially due to the
laboratory data. We compared twenty empirical inaccessibility of samples during the early project
correlations for the determination of the solution stage when only surface tests are carried out.
gas-oil ratio, oil formation-volume-factor and Hence, the lack of laboratory measurements led
under-saturated viscosity (μo) for oil samples from practitioners to use empirical correlations for
Brazilian pre-salt reserves originally evaluated by estimating the volumetric behavior of reservoir
130
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
a
Table 1. Examples of PVT correlations .
fluids in terms of different pressure and visualization of the first bubbles or graphically in a
temperature conditions (Asadisaghandi & pressure to volume (PV) diagram, as shown in
Tahmasebi, 2011; Karimnezhad et al., 2014). Said Figure 1. The PV diagram is obtained through
models have been used widely for estimating Constant Composition Expansion experiments. Red
reservoir fluid properties, comparing data with and blue curves correspond to the two-phase and
laboratory results to identify associated errors and one-phase regions, respectively, whereas the
determine the quality of samples. inflection point represents the value of Pb.
Numerous studies regarding the foundation of The oil formation volume factor, in turn, refers
PVT correlations have been developed, among to the total volume of oil required to produce one
which some examples are disclosed in Table 1. We barrel under standard conditions, i.e. at 60° F and
note that the majority of the correlations published
so far refer to North American fields, where the oil
industry started to develop since the first oil well
drilling in 1859. We can also observe that bubble
point pressure, solution gas-oil ratio, and oil
formation-volume-factor received the most
attention due to the fact that, according to
Olatunji, Selamat, & Raheem (2011), “they are the
most essential factors in reservoir and production
computations”.
131
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
V
SC
g
R = (2)
Vo
s SC
132
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
for a wide variety of mixtures from different meters below sea level; in ultra-deep water (about
regions (Mansour et al., 2013; Ahmadi et al., 2,000 meters); and under a thick salt layer (up to
2019). However, according to Olatunji et al. 2,000 meters) (Magalhães & Domingues, 2014;
(2011), EOS models require well-defined Kutas, 2015), as illustrated in Figure 4.
composition data and depend heavily on fluid
characteristics and tuning procedures. To The total oil volume in such an area is estimated
overcome such drawbacks, numerous practitioners at about 80 billion barrels, which could make Brazil
have resorted to other computational strategies one of the world’s biggest holders of oil reserves
that may include, but are not limited to, artificial (Magalhães & Domingues, 2014). In fact, the pre-
neural networks (Asadisaghandi & Tahmasebi, salt region is comprised of various accumulations,
2011; Talebi et al., 2014; Mahdiani & Kooti, 2016); among which the Campos Basin stands out as the
support vector machine (Rafiee-Taghanaki et al., first discovery and the main sedimentary region
2013; Khamis & Fattah, 2019); and fuzzy logic from the Brazilian coastline to be explored. It has a
systems (Olatunji et al., 2011; Salehinia et al., total area of approximately 100,000 square
2016) among others. Nevertheless, said computer kilometers and currently accounts for 42% of the
techniques present some shortcomings related to total oil production in Brazil (Pré-Sal Petróleo S.A.,
instability and inconsistency when handling 2019).
uncertainties.
The exploration of the pre-salt region is not only
2.2 Brazilian Pre-salt unprecedented but also challenging due to the
unfavorable natural conditions. To overcome
In late 90s, the Brazilian oil industry started to technical difficulties, it is a must to employ highly
track down new reserves to perpetuate its experienced professionals, use advanced
production levels. With the development of new technologies, and acquire a vast practical
exploration technologies (e.g. 4D seismic) and after knowledge about the area. In this regard, scientific
sedulous searching operations, the pre-salt layer oil studies play an important role in providing the
was discovered in 2006-2007 by Petróleo Brasileiro information needed to develop new strategies
S.A. (Petrobras) (Johann & Monteiro, 2016). Such towards progress.
milestone has impacted the Country’s economic
and political spheres over the past decade due to
its potential to lead the petroleum industry
towards a significant development. 3. METHODS
The pre-salt layer is located in the Brazilian Various studies regarding the development
exclusive economic zone. It encompasses several and/or evaluation of PVT correlations have been
oil reserves situated up to 300 km from the coast published, which evidence the significance of these
with different depths between 5,000 and 6,000 data to the oil industry. However, most of these
133
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
b
Table 2. Range of applicability for solution gas-oil ratio correlations .
Author T (° F) P (psi) γo γg
Standing (1947) 100-258 20-1425 0.72-0.96 0.59-0.95
Glasø (1980) 80-280 90-2637 0.79-0.92 0.65-1.28
Vazquez and Beggs (1980) 75-294 0-2199 0.74-0.96 0.51-1.35
Ostermann and Owolabi (1983) 122-180 515-1802 0.84-0.90 0.85-1.09
Al-Marhoun (1985) 74-240 26-1602 0.80-0.94 0.75-1.37
Obomanu and Okpobiri (1987) 123-190 73-2529 0.81-0.97 0.56-0.93
Farshad et al. (1992) 95-260 6-1645 0.80-0.95 0.66-1.7
Hasan et al. (1993) - - - -
Macary and El-Batanoney (1993) 130-290 200-1200 0.82-0.90 0.7-1.00
Petrosky and Farshad (1993) 114-288 217-1406 0.80-0.96 0.58-0.85
Kartoatmodjo and Schmidt (1994) 75-320 0-2890 0.74-0.97 0.38-1.17
Elsharkawy and Alikhan (1997) 130-250 39-1586 0.81-0.95 0.66-1.27
Levitan and Murtha (1999) 60-250 - 0.70-1.00 0.60-0.90
Al-Shammasi (2001) 74-342 32-7127 0.73-1.03 0.51-3.44
Dindoruk and Christman (2004) 117-276 926-12230 0.83-0.87 0.76-1.03
Hemmati and Kharrat (2007) 78-290 348-5156 0.79-0.94 0.52-1.42
Mazandarani and Asghari (2007) 100-250 1000-6000 0.84-0.90 0.87-1.17
Ikiensikimama and Ajienka (2012) 122-264 67-6560 0.77-0.97 0.56-1.29
Arabloo et al. (2015) 74-361 73-6629 0.75-1.03 0.52-3.44
Jarrahian et al. (2015) 68-330 83-6875 0.77-0.97 0.60-1.53
b
T: Temperature. P: Pressure. γo: Oil gravity. γg: Gas gravity.
c
Table 3. Range of applicability for oil formation volume factor correlations .
Author T (° F) Rs (scf/STB) γo γg
Standing (1947) 100-258 20-1425 0.72-0.96 0.59-0.95
Glasø (1980) 80-280 90-2637 0.79-0.92 0.65-1.28
Vazquez and Beggs (1980) 75-294 0-2199 0.74-0.96 0.51-1.35
Al-Marhoun (1985) 74-240 26-1602 0.80-0.94 0.75-1.37
Obomanu and Okpobiri (1987) 123-190 73-2529 0.81-0.97 0.56-0.93
Abdul-Majeed and Salman (1988) 75-290 0-1664 0.74-1.00 -
Labedi (1990) 100-306 13-3533 0.79-0.86 0.70-1.47
Dokla and Osman (1992) 190-275 181-2266 0.82-0.89 0.80-1.29
Farshad et al. (1992) 95-260 6-1645 0.80-0.95 0.66-1.7
Macary and El-Batanoney (1993) 130-290 200-1200 0.82-0.90 0.7-1.00
Omar and Todd (1993) 125-280 142-1440 0.77-0.89 0.61-1.32
Petrosky and Farshad (1993) 114-288 217-1406 0.80-0.96 0.58-0.85
Kartoatmodjo and Schmidt (1994) 75-320 0-2890 0.74-0.97 0.38-1.17
Almehaideb (1997) 190-306 128-3871 0.79-0.87 0.75-1.12
Elsharkawy and Alikhan (1997) 130-250 39-1586 0.81-0.95 0.66-1.27
El-Banbi et al. (2006) 186-312 0-8280 0.74-0.85 0.67-1.14
Karimnezhad et al. (2014) 59-306 9-3299 0.77-1.03 0.62-1.53
Sulaimon (2014) 125-280 142-1440 0.77-0.89 0.04-1.32
Torabi et al. (2014) 60-140 - 0.80-1.03 -
Mahdiani and Kooti (2016) 100-280 104-1439 0.77-0.90 0.61-1.32
c
T: Temperature. Rs: Solution gas-oil ratio. γo: Oil gravity. γg: Gas gravity.
studies are based on restricted criteria that usually ones best represent the characteristics of the
depend on specific properties of the given region Campos Basin reserves of the Brazilian pre-salt
where the data were retrieved from (Rafiee- region. The models for the estimation of Rs, Bo, and
Taghanaki et al., 2013). Therefore, in this study, we μo studied on the present study are disclosed in
compared different correlations to assert which Appendices A, B, and C, respectively.
134
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
d
Table 4. Range of applicability for undersaturated oil viscosity correlations .
e
Table 5. Data ranges considered in this work (according to Elias & Trevisan, 2016) .
Parameter T (° F) P (psi) γo γg
Rs 68-204 711-5759.1 0.71-0.84 0.93-1.13
Bo 68-204 711-5759.1 0.71-0.84 0.93-1.13
μo 68-204 5759.1-9243.0 0.71-0.84 0.93-1.13
e
T: Temperature. P: Pressure. γo: Oil gravity. γg: Gas gravity.
Yet, one should note that all correlations may The disclosed correlations were compared to
be applicable only depending on specific the experimental data presented by Elias and
parameters. For instance, the equations evaluated Trevisan (2016), who recombined dead oil and
in this paper can only be used for the ranges of synthetic gas to obtain an original pre-salt live oil
temperature, pressure, oil and gas gravities sample. Although the authors stated that they had
disclosed in Tables 2, 3, and 4, respectively. This is analyzed oil samples found in Brazilian pre-salt
due to the fact that such prediction models were reserves in the region of Campos Basin, it seems
originally developed through experimental data that they have opted for not providing further
that were measured under those specific information regarding the respective oil fields due
conditions. The same applies to Elias and to confidentiality purposes. However, one may
Trevisan’s (2016) study, which was based on the infer that the experiments were carried out using
data ranges shown in Table 5. In this regard, it is oil samples that are representative of such a
expected that correlations using ranges of region.
parameters that are different from the respective
ranges considered by said authors may provide The qualitative comparison was carried out by
inaccurate or erroneous results. In addition to said means of crossplots regressions, whereas the
tables, Figures 5, 6, and 7 illustrate the comparison quantitative analysis was performed through the
between the validity ranges of the different calculation of the Average Absolute Relative Error
correlations, and the ranges corresponding to the (AARE), Root Mean Square Error (RMSE), and
experimental data used by the referenced authors. Standard Error of Estimate (SEE), as shown in
135
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
0.5
1 n 2
RMSE= X calc -X exp (4)
n i=1 i
0.5
n
SEE= X calc -X exp n v 1
2
(5)
i=1 i
136
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
an AARE, a RMSE, and a SEE as low as 0.4956%, accuracy of Vazquez and Beggs (1980), Al-
0.0116, and 0.0174, respectively. Marhoun (1985), Farshad et al. (1992), and
Almehaideb’s (1997) models can be associated
An analysis of Table 7 points out that models with their respective data ranges, which
presented by Al-Marhoun (1985), Petrosky and encompass the ranges considered by Elias and
Farshad (1993), Vazquez and Beggs (1980), Trevisan’s (2016) work. The ranges considered by
Farshad et al. (1992), and Almehaideb (1997) also Obomanu and Okpobiri (1987), Macary and El-
showed good performances. Once more, Obomanu Batanoney (1993), and Elsharkawy and Alikhan
and Okpobiri’s (1987) equation showed the worst (1997) are significantly different than those studied
prediction of performance, since it has the highest by Elias and Trevisan (2016) and, therefore, are
AARE (66.2463%), RMSE (1.0617), and SEE not representative of the latter’s data.
(1.5926). Other correlations that may be deemed
quite inaccurate for Bo estimation are Macary and Finally, the results of the μo correlations are
El-Batanoney’s (1993) (AARE of 50.3572%, RMSE of disclosed in Table 8. Based on these results we can
0.7562, and SEE of 1.1343) and Elsharkawy and state that Beal’s (1946) model has the best
Alikhan’s (1997) (AARE of 51.9248%, RMSE of performance, having an AARE, a RMSE, and a SEE
0.8703, and SEE of 1.3055). as low as 0.4557%, 0.0034, and 0.0051,
respectively. In addition, Almehaideb (1997) and
Comparably to the analysis of Rs models, the Al-Marhoun’s (2015) correlations also proved to be
137
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
representative of Elias and Trevisan’s (2016) data, for μo estimation are Hossain et al. (2005),
since their statistical errors were quite low. Isehunwa et al. (2006), and Abedini et al.’s (2010).
Elsharkawy and Alikhan’s (1999) equation,
however, can be deemed the worst model for One can verify that, in the case of μo, the
predicting the under-saturated oil viscosity of the performance of PVT models does not seem to be
Campos Basin samples due to its significant errors strictly dependent on their respective data ranges.
(AARE of 16.2215%, RMSE of 0.1468, and SEE of Figure 7 shows that the aforementioned
0.2203). Other correlations that, due to large correlations, including the best ones, have different
errors, may not be considered as representatives ranges from the ones proposed by Elias and
138
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
Trevisan (2016). It is worth highlighting that, in this references evaluated data sets with compositions
article, we aimed at comparing the behavior of similar to the ones found for Elias and Trevisan’s
generic PVT correlations when estimating fluid (2016) samples.
properties of the Campos Basin fluid samples,
The qualitative analysis for the comparison of
instead of calibrating the parameters and/or
Rs, Bo, and μo correlations with Elias and Trevisan’s
developing specific models.
(2016) data was then performed through the
It is worth pointing out that the correlations crossplots illustrated in Figures 8, 9, and 10,
that best predicted Rs for the Campos Basin oil respectively. Regarding the solution gas-oil ratio,
samples were developed originally for different most models resulted in under-predicted data,
fields. For instance, Al-Shammasi (2001), Arabloo while Farshad et al. (1992), Macary and El-
et al. (2015), and Jarrahian et al. (2015) carried out Batanoney (1993), Ikiensikimama and Ajienka
their evaluation on the basis of global data sets, (2012), and Arabloo et al. (2015) over-estimated
while Elsharkawy and Alikhan (1997) evaluated the values for higher pressures. In relation to the
samples from Kuwait. In relation to Bo, Al-Marhoun oil formation volume factor, in turn, Abdul-Majeed
(1985) and Almehaideb (1997) analyzed and Salman (1988), Kartoatmodjo and Schmidt
experimental data from the Middle East region, (1994), Omar and Todd (1993), El-Banbi et al.
whereas Petrosky and Farshad (1993) based their (2006), and Sulaimon (2014) under-predicted the
analysis on samples obtained in the Gulf of Mexico experimental data, whereas Dokla and Osman
area. Farshad et al. (1992) and Vazquez and Beggs (1992), Elsharkawy and Alikhan (1997), Obomanu
(1980), in turn, used oil samples from Colombian and Okpobiri (1987), Macary and El-Batanoney
fields and global data sets, respectively. Finally, (1993), and Mahdiani and Kooti (2016) presented
regarding the predictions of μo, Beal’s (1946) over-estimated results. Finally, with regard to
model was based on data retrieved in the United under-saturated oil viscosity, Abdul-Majeed et al.
States, while Almehaideb (1997) and Al-Marhoun (1990), Kartoatmodji and Schmidt (1991), Labedi
(2015) developed their equations on the basis of (1992), and Hemmati-Sarapardeh et al. (2013)
samples from the United Arab Emirates. This provided under-predicted results, while Kouzel
corroborates Danesh’s (1998) statement (1965), Khan et al. (1987), De Ghetto et al. (1994),
concerning the fact that petroleum properties do Petrosky and Farshad (1995), Elsharkawy and
not depend on the reserve location itself, but Alikhan (1999), Hossain et al. (2005), Isehunwa et
instead on fluid composition and reservoir al. (2006), and Abedini et al.’s (2010) models
conditions. Hence, it is possible that said resulted in over-predicted data.
139
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
140
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
141
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
142
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
143
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
144
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
145
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
In accordance with the quantitative results To the best of knowledge, most of the
disclosed in Tables 6, 7, and 8, we verify that Al- published contributions encompassing the
Marhoun (1985), Al-Shammasi (2001) and Beal’s presence of high CO2 contents in oil mixtures refer
(1946) correlations clearly are almost perfectly fit to equation of state modeling (Elias et al., 2014),
onto their respective equality reference lines. assessment of CO2-oil miscibility pressure
Therefore, they can be considered as the best (Shokrollahi et al., 2015), and development of
models for estimating Rs, Bo, and μo of oils from the specific CO2-dead oil viscosity and solubility models
Campos Basin considered by Elias and Trevisan (e.g. Welker & Dunlop, 1963; Simon & Graue,
(2016), respectively. Although current practices 1965; Barclay & Mishra, 2016). In this context,
frequently use computational methods for future studies should focus on considering the CO2
performing calculations, simple equation forms composition either as an independent variable in
may be preferred. In this regard, Elsharkawy and empirical correlations or as a validity range within
Alikhan’s (1997) model may be selected for Rs which such models could be well-represented.
estimation, while Almehaideb’s (1997) correlation Nevertheless, compositional modeling and
may be used to provide good and easy estimates simulation would be preferable since the accuracy
for Bo. In terms of μo correlations, Beal’s (1946) of PVT correlations depend highly on the specific
model may be already deemed the simplest. geographical regions from where the oil mixtures
are retrieved, which may render them fairly
It is noteworthy that most of the correlations laborious work due to the necessity of acquiring an
studied in this paper presented large statistical extensive amount of field data.
errors because the analysis of the present work
was carried out in a relatively limited range of data.
Such errors could be reduced through the
development of specific models for pre-salt fluids. 5. CONCLUSIONS
More specifically, the development of models for
oils from the Campos Basin region – which would Bubble point pressure, solution gas-oil ratio,
be performed by calibrating the parameters of the and oil formation volume factor are interpreted as
respective equations. However, a more substantial being the most relevant parameters for reservoir
amount of experimental data would be necessary. engineering and field development. Such PVT data
may be acquired either from laboratory analysis or,
In addition, a possible reason for obtaining such in case of its absence, through empirical
large errors may be related to the high contents of correlations.
CO2 in pre-salt reserves, which vary from about 8%
(e.g. Tupi reserve, according to Beltrão et al., 2009) Numerous models have been proposed for
to approximately 44% (e.g. Libra field, according to different oil types and reserves around the world.
Arinelli et al., 2015). The presence of carbon However, there is a lack of studies regarding the
dioxide in oil reservoirs influences strongly the Brazilian pre-salt region. Specifically, since it has
phase equilibria of such mixtures since CO2 is reshaped the Brazil’s economy and has the
soluble with crude oil (much more than other potential of making it one of the world’s biggest
conventional gases), especially when the former’s holders of oil reserves. Hence, we compared
density increases due to compression. As discussed twenty empirical correlations for the
by Mansour et al. (2019), high-pressure CO2 determination of the under-saturated oil viscosity,
dissolves in the oil, swells it, and reduces its solution gas-oil ratio, and oil formation volume
viscosity, increasing dramatically the gas solubility factor of oil samples from the Campos Basin
and the drive energy available to reservoir reserves in the Brazilian pre-salt region originally
production. For this particular reason, and evaluated by Elias & Trevisan (2016). The results
considering environmental issues, re-injection of obtained by the authors were tested against each
CO2 has become an important strategy for correlation to determine the model with the best
enhanced oil recovery. It is therefore safe to say prediction of performance.
that PVT correlations that do not account for
The qualitative comparison was performed by
higher carbon dioxide compositions tend to
plotting the data in regression crossplots, whereas
estimate over-predicted μo data and under-
the quantitative analysis was carried out through
predicted Rs values.
the calculation of the Average Absolute Relative
146
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
Error (AARE), Root Mean Square Error (RMSE), and Volume of oil and dissolved gas under
(VR)T,P
Standard Error of Estimate (SEE). From our results, reservoir conditions
and with regard to the Campos Basin fluids, Al- Xcalc Calculated value
Shammasi’s (2001) correlation proved to be the
best estimation model for solution gas-oil ratio, Xexp Experimental result
whereas Al-Marhoun (1985) and Beal’s (1946)
correlations were deemed the most accurate for Greek symbols
the prediction of oil formation volume factor and
under-saturated oil viscosity, respectively. Oil API gravity
γAPI
Although the analysis of the present work was
γg Gas gravity
carried out in terms of a relatively limited range of
data, it is believed that the purpose of providing a γo Oil gravity
statistical comparison and defining the known μo Under-saturated oil viscosity
correlations for predicting PVT properties of pre-
salt fluids from the Campos Basin area could be μob Bubble-point oil viscosity
appropriately achieved. Nevertheless, future works μod Dead oil viscosity
should aim at gathering a more substantial amount
of laboratory data to develop a specific model for
such fluids.
ABBREVIATIONS
AARE Average Absolute Relative Error
ACKNOWLEDGMENTS API American Petroleum Institute
This study was financed in part by the Coordenação CAPES Coordenação de Aperfeiçoamento de
de Aperfeiçoamento de Pessoal de Nível Superior - Pessoal de Nível Superior
Brasil (CAPES) - Finance Code 001.
PVT Pressure-Volume-Temperature
RMSE Root Mean Square Error
Bo Oil formation-volume-factor
Bob Bubble-point oil formation-volume-factor 6. REFERENCES
Co Under-saturated oil compressibility
Abedini, R.; Abedini, A.; Yakhfrouzan, N. E. A
n Number of experimental data points new correlation for prediction of undersaturated
P Pressure crude oil viscosity. Petroleum & Coal, v.52, p. 50-
55, 2010.
Pb Bubble-point pressure
Abdul-Majeed, G.H.; Salman, N.H. An empirical
Psep Separator pressure
correlation for oil FVF prediction. Journal of
Rs Solution gas-oil ratio Canadian Petroleum Technology, v.27, p. 118-122,
1988. https://doi.org/10.2118/88-06-10
Rsb Bubble-point gas-oil ratio
SC Standard Conditions Abdul-Majeed, G.H.; Kattan, R.R.; Salman, N.H.
New Correlation for Estimating the Viscosity of
T Temperature Undersaturated Crude Oils. Journal of Canadian
Tsep Separator temperature Petroleum Technology, v.29, p. 80-85, 1990.
https://doi.org/10.2118/90-03-10
SC Volume of gas produced under standard
(Vg)
conditions
SC
(Vo) Volume of oil under standard conditions
147
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
Ahmed, T. Reservoir engineering handbook. Barclay, T.H.; Mishra, S. New correlations for
Fourth ed. Oxford: Gulf Professional Publishing, CO2-Oil solubility and viscosity reduction for light
2010. 1524p. oils. Journal of Petroleum Exploration and
Production Technology, v.6, p. 815-823, 2016.
Al-Marhoun, M.A. Black oil property
https://doi.org/10.1007/s13202-016-0233-y
correlations - State of the art. In: SPE Middle East
Oil & Gas Show and Conference, MEOS 2015, Beal, C. The viscosity of air, water, natural gas,
Manama, p. 1-12, 2015. crude oil and its associated gases at oil field
https://doi.org/10.2118/172833-MS temperature and pressures. Transactions, v.165, p.
94-115, 1946. https://doi.org/10.2118/946094-G
Al-Marhoun, M.A. PVT correlations for Middle
East crude oils. In: SPE Middle East Oil Technical Beltrão, R.L.C.; Sombra, C.L.; Lage, A.C.V.M.;
Conference & Exhibition, MEOTCE 1985, Manama, Netto, J.R.F.; Henriques, C.C.D. Challenges and new
1985. technologies for the development of the pre-salt
cluster, Santos Basin, Brazil. In: Offshore
Al-Shammasi, A.A. A Review of Bubble Point
Technology Conference, OTC 2009, Houston, p. 1-
Pressure and Oil Formation Volume Factor
11, 2009. https://doi.org/10.4043/OTC-19880-MS
Correlations. SPE Reservoir Evaluation &
Engineering, v.4, p. 146-160, 2001. Bergman, D.F.; Sutton, R.P. Undersaturated Oil
https://doi.org/10.2118/71302-PA Viscosity Correlation for Adverse Conditions. SPE
Almehaideb, R.A. Improved PVT correlations for Annual Technical Conference and Exhibition, ATCE
UAE crude oils. In: Middle East Oil Show and 2006, San Antonio, p. 1-20, 2006.
Conference, MEOS 1997, Manama, p. 109-120, Caixerio, E. Correlation of reservoir properties,
1997. https://doi.org/10.2118/37691-MS Miranga field. M.Sc. Report, Stanford University,
Alqahtani, N.B.; AlQuraishi, A.A.; Al-Baadani, W. 1976.
New correlations for prediction of saturated and Danesh, A. PVT and phase behaviour of
undersaturated oil viscosity of Arabian oil fields. petroleum reservoir fluids. Vol. 47, First ed.
Journal of Petroleum Exploration and Production Amsterdam: Elsevier Science, 1998. 400p.
Technology, v.8, p. 205-216, 2018.
https://doi.org/10.1007/s13202-017-0332-4 De Ghetto, G.; Paone, F.; Villa, M. Reliability
Analysis on PVT Correlations. In: SPE European
Arabloo, M.; Amooiea, M.A.; Hemmati, S.A.;
Petroleum Conference, EPC94, London, p. 375-393,
Ghazanfari, M.H.; Mohammadi, A.H. Application of
1994. https://doi.org/10.2118/28904-MS
constrained multi-variable search methods for
prediction of PVT properties of crude oil systems. Dindoruk, B.; Christman, P.G. PVT Properties
Fluid Phase Equilibria, v.363, p. 121-130, 2015. and Viscosity Correlations for Gulf of Mexico Oils.
https://doi.org/10.1016/j.fluid.2013.11.012 In: SPE Annual Technical Conference and
Exhibition, ATCE 2004, New Orleans, p. 427-437,
Arinelli, L.O.; de Medeiros, J.L.; Araújo,O.Q.
2004. https://doi.org/10.2118/89030-PA
Performance Analysis and Comparison of
Membrane Permeation Versus Supersonic Dokla, M.E.; Osman, M.E. Correlation of PVT
Separators for CO2 Removal From a Plausible properties for UAE crudes. SPE Formation
Natural Gas of Libra Field, Brazil. In: Offshore Evaluation, v.7, p. 41-46, 1992.
Technology Conference, OTC Brasil 2015, Rio de https://doi.org/10.2118/20989-PA
Janeiro, p. 1-12, 2015. https://doi.org/10.4043/26164-
MS
148
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
El-Banbi, A.H.; Fattah, K.A.; Sayyouh, M.H. New Hasan, M.; Lestari; Inawati. PVT Correlations for
modified black-oil PVT correlations for Gas Indonesian Crude Oils. In: 22nd Annual Convention
condensate and volatile oil fluids. In: SPE Annual of the Indonesian Petroleum Association, 22nd
Technical Conference and Exhibition, ATCE 2006, Annual Convention Proceedings, p. 237-254, 1993.
San Antonio, p. 1-6, 2006.
https://doi.org/10.2118/102240-MS Hemmati, M.N.; Kharrat, R. A correlation
approach for prediction of crude oil PVT properties.
Elam, F.M. Prediction of bubble point pressure In: SPE Middle East Oil and Gas Show and
and formation volume factors from field data. Conference, MEOS 2007, Manama, 2007.
M.Sc. Thesis, Texas A&M University, 1957. https://doi.org/10.2118/104543-MS
Elias, A.; Mello, S.F.; Schiozer, D.J.; Trevisan,
O.V. PVT analysis process of pre-salt similar oil Hemmati-Sarapardeh, A.; Khishvand, M.; Naseri,
associated with compositional simulation. In: Rio A.; Mohammadi, A.H. Toward reservoir oil viscosity
Oil & Gas Expo and Conference 2014, p.1-9, 2014. correlation. Chemical Engineering Science, v.90, p.
53-68, 2013. https://doi.org/10.1016/j.ces.2012.12.009
Elias, A.; Trevisan, O.V. An experimental
investigation on phase behavior of a light oil and Hossain, M.S.; Sarica, C.; Zhang, H.-Q.
CO2. Journal of Petroleum Science and Assessment and development of heavy-oil viscosity
Engineering, v.145, p. 22-33, 2016. correlations. In: SPE/PS-CIM/CHOA International
https://doi.org/10.1016/j.petrol.2016.03.006 Thermal Operations and Heavy Oil Symposium,
ITOHOS 2005, Calgary, p. 1-9, 2005.
Elsharkawy, A.M.; Alikhan, A.A. Models for https://doi.org/10.2118/97907-MS
predicting the viscosity of Middle East crude oils.
Fuel, v.78, p. 891-903, 1999. Ikiensikimama, S.S.; Ajienka, J.A. Impact of PVT
https://doi.org/10.1016/S0016-2361(99)00019-8 correlations development on hydrocarbon
accounting: The case of the Niger Delta. Journal of
Elsharkawy, A.M.; Alikhan, A.A. Correlations for
Petroleum Science and Engineering, v.81, p. 80-85,
predicting solution gas–oil ratio, oil formation
2012. https://doi.org/10.1016/j.petrol.2011.12.017
volume factor and undersaturated oil
compressibility. Journal of Petroleum Science and Isehunwa, O.S.; Olamigoke, O.; Makinde, A.A. A
Engineering, v.17, p. 291-302, 1997. Correlation to Predict the Viscosity of Light Crude
https://doi.org/10.1016/S0920-4105(96)00075-7 Oils. In: SPE Nigeria Annual International
Elsharkawy, A.M.; Elgibaly, A.A.; Alikhan, A.A. Conference and Exhibition, NAICE2006, Abuja, p. 1-
Assessment of the PVT correlations for predicting 7, 2006. https://doi.org/10.2118/105983-MS
the properties of Kuwaiti crude oils. Journal of
Petroleum Science and Engineering, v.13, p. 219- Jarrahian, A.; Moghadasi, J.; Heidaryan, E.
232, 1995. https://doi.org/10.1016/0920- Empirical estimating of black oils bubble point
4105(95)00012-7 (saturation) pressure. Journal of Petroleum
Science and Engineering, v.126, p. 69-77, 2015.
Farshad, F.F.; LeBlance, J.L.; Garbeer, J.O.; https://doi.org/10.1016/j.petrol.2014.12.004
Osorio, J.G. Empirical PVT Correlations for
Colombian Crudes. In: SPE Latin America/Caribbean Johann, P.; Monteiro, R.C. Geophysical
Petroleum Engineering Conference, LACPEC 1992, Reservoir Characterization and Monitoring at
Port of Spain, p. 311-320, 1992. Brazilian Pre-Salt Oil Fields. In: Offshore Technology
Conference, 2016, OTC 2016, Houston, p. 1-27,
Gharbi, R.; Elsharkawy, A.M. Predicting the 2016. https://doi.org/10.4043/27246-MS
Bubble-Point Pressure and Formation-Volume-
Factor of Worldwide Crude Oil Systems. Petroleum Karimnezhad, M.; Heidarian, M.; Kamari, M.;
Science and Technology, v.21, p. 53-79, 2003. Jalalifar, H. A new empirical correlation for
https://doi.org/10.1081/LFT-120016921 estimating bubble point oil formation volume
factor. Journal of Natural Gas Science and
Glasø, O. Generalized pressure-volume- Engineering, v.18, p. 329-335, 2014.
temperature correlations. Journal of Petroleum https://doi.org/10.1016/j.jngse.2014.03.010
Technology, v.32, p. 785-795, 1980.
https://doi.org/10.2118/8016-PA
149
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
Kartoatmodjo, T.; Schmidt, Z. Large data bank Levitan, L.L.; Murtha, M. New Correlations
improves crude oil physical property correlations. Estimate Pb, FVF. Oil & Gas Journal, v.97, p. 70,
Oil & Gas Journal, p. 51-55, 1994. 1999. https://doi.org/10.1016/S0026-0576(99)80583-2
Kartoatmodjo, T.; Schmidt, Z. New Correlations Macary, S.; El-Batanoney, M. Derivation of PVT
For Crude Oil Physical Properties. SPE paper 23556, correlations for the Gulf of Suez crude oils. Journal
1991. of the Japan Petroleum Institute, v.36, p. 472-478,
1993. https://doi.org/10.1627/jpi1958.36.472
Katz, D.L. Prediction of the Shrinkage of Crude
Oils. Drilling and Production Practice. p. 137-147, Magalhães, A.S.; Domingues, E.P. Blessing or
1942. curse: Impacts of the Brazilian Pre-Salt oil
exploration. EconomiA, v.15, p. 343-362, 2014.
Khamis, M.A.; Fattah, K.A. Estimating oil–gas https://doi.org/10.1016/j.petlm.2015.12.001
ratio for volatile oil and gas condensate reservoirs:
artificial neural network, support vector machines Mahdiani, M.R.; Kooti, G. The most accurate
and functional network approach. Journal of heuristic-based algorithms for estimating the oil
Petroleum Exploration and Production formation volume factor. Petroleum, v.2, p. 40-48,
Technology, v.9, p. 573-582, 2019. 2016. https://doi.org/10.1016/j.petlm.2015.12.001
https://doi.org/10.1007/s13202-018-0501-0
Management Centre Europe. Oil & Gas
Khan, S.A.; Al-Marhoun, M.A.; Duffuaa, S.O.; Industry. Available at: https://mce.eu/oil-gas/.
Abu-Khamsin, S.A. Viscosity Correlations for Saudi Accessed on: 29 October 2018.
Arabian Crude Oils. In: SPE Middle East Oil Show,
MEOS 1987, Manama, p. 251-258, 1987. Mansour, E.M.; Al-Sabagh, A.M.; Desouky, S.M.;
https://doi.org/10.2118/15720-MS Zawawy, F.M.; Ramzi, M. A laboratory investigation
of carbon dioxide-enhanced oil recovery by
Knopp, C.R.; Ramsey, L.A. Correlation of oil focusing on CO2-oil physical properties. Egyptian
formation volume factor and solution gas–oil ratio. Journal of Petroleum, v.28, p. 21-26, 2019.
Journal of Petroleum Technology, v.12, p. 27–29, https://doi.org/10.1016/j.ejpe.2018.10.004
1960. https://doi.org/10.2118/1433-G
Mansour, E.M.; Farag, A.B.; El-Dars, F.S.;
Kouzel, B. How pressure affects liquid viscosity. Desouky, S.M.; Batanoni, M.H.; Mahmoud, M.R.M.
Hydrocarbon Processing & Petroleum Refiner, Predicting PVT properties of Egyptian crude oils by
v.44, p. 120, 1965. a modified Soave–Redlich–Kowng equation of
state. Egyptian Journal of Petroleum, v.22, p. 137-
Kutas, D.T. A Study of the Applicability of 148, 2013. https://doi.org/10.1016/j.ejpe.2012.09.005
Bourgoyne & Young ROP Model and Fitting
Reliability through Regression. In: International Mazandarani, M.T.; Asghari, S.M. Correlations
Petroleum Conference, IPTC 2015, Doha, p. 1-14, for predicting solution gas-oil ratio, bubble point
2015. https://doi.org/10.2523/IPTC-18521-MS pressure, and oil formation volume factor at
bubble Point of Iran crude oils. In: 6th European
Labedi, R. Improved correlations for predicting Congress of Chemical Engineering, ECCE-6,
the viscosity of light crudes. Journal of Petroleum Copenhagen, p. 1-8, 2007.
Science and Engineering, v.8, p.221-234, 1992.
https://doi.org/10.1016/0920-4105(92)90035-Y Nnabuo, N.N.; Okafor, I.S.; Ubani, C.E.
Interpretation of Laboratory PVT Analysis Result (A
Labedi, R. Use of production data to estimate Case Study of a Niger Delta Field). In: SPE Nigeria
volume factor density and compressibility of Annual International Conference and Exhibition,
reservoir fluids. Journal of Petroleum Science and NAICE2014, Lagos, p. 1-10, 2014.
Engineering, v.4, p. 357-390, 1990. https://doi.org/10.2118/172412-MS
https://doi.org/10.1016/0920-4105(90)90034-Z
Obomanu, D.A.; Okpobiri, G.A. Correlating the
Lasater, J.A. Bubble point pressure correlation. PVT properties of Nigerian crudes. Journal of
Journal of Petroleum Technology, v.10, p. 65-67, Energy Resources Technology, v.109, p. 214-217,
1958. https://doi.org/10.2118/957-G 1987. https://doi.org/10.1115/1.3231349
150
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
Olatunji, S.O.; Selamat, A.; Raheem, A.A.A. Rasouli, H.; Rashidi, F.; Ebrahimian, A.
Predicting correlations properties of crude oil Estimating the Bubble Point Pressure and
systems using type-2 fuzzy logic systems. Expert Formation Volume Factor of Oil Using Artificial
Systems with Applications, v.38, p. 10911-10922, Neural Networks. Chemical Engineering &
2011. https://doi.org/10.1016/j.eswa.2011.02.132 Technology, v.31, p. 493-500, 2008.
https://doi.org/10.1002/ceat.200700434
Omar, M.I.; Todd, A.C. Development of new
modified black oil correlations for Malaysian Salehinia, S.; Salehinia, Y.; Alimadadi, F.; Sadati,
crudes. In: SPE Asia Pacific Oil and Gas Conference, S.H. Forecasting density, oil formation volume
APOGCE 1993, Singapore, p. 211-219, 1993. factor and bubble point pressure of crude oil
https://doi.org/10.2118/25338-MS systems based on nonlinear system identification
approach. Journal of Petroleum Science and
Ostermann, R.D.; Owolabi, O.O. Correlations for Engineering, v.147, p. 47-55, 2016.
the reservoir fluid properties of Alaskan crudes. In: https://doi.org/10.1016/j.petrol.2016.05.008
SPE California Regional Meeting, CRM 1983,
Ventura, p. 357-366, 1983. Shokrollahi, A.; Tatar, A.; Safari, H. On accurate
https://doi.org/10.2118/11703-MS determination of PVT properties in crude oil
systems: Committee machine intelligent system
Pedersen, K.S.; Christensen, P.L.; Shaikh, J.A. modelling approach. Journal of the Taiwan
Phase behavior of petroleum reservoir fluids. Institute of Chemical Engineers, v. 55, p. 17-26,
Second ed. New York: CRC Press, 2015. 2015. https://doi.org/10.1016/j.jtice.2015.04.009
https://doi.org/10.1201/b17887
Simon, R.; Graue, D.J. Generalized Correlations
Petrosky, G.E.; Farshad, F.F. Viscosity for Predicting Solubility, Swelling and Viscosity
correlations for Gulf of Mexico crude oils. In: SPE Behavior of CO2-Crude Oil Systems. Journal of
Production Operations Symposium, SPEPOS 1995, Petroleum Technology, v.17, p. 102-106, 1965.
Oklahoma City, p. 249-258, 1995. https://doi.org/10.2118/917-PA
https://doi.org/10.2118/29468-MS
Standing, M.B. A Pressure Volume Temperature
Petrosky, G.E.; Farshad, F.F. Pressure volume Correlation for Mixture of California Oils and Gases.
temperature correlations for Gulf of Mexico crude Drilling and Production Practice, p. 275-287, 1947.
oils. SPE Reservoir Evaluation & Engineering, v.1,
p. 416-420, 1993. https://doi.org/10.2118/51395-PA Sulaimon, A.A.; Ramli, N.; Adeyemi, B.J. New
Correlation for Oil Formation Volume Factor. In:
Potsch, K.; Toplack, P.; Gumpenberger, T. A SPE Nigerian Annual International Conference and
Review and Extension of Existing Consistency Exhibition, NAICE2014, Lagos, p. 1-6, 2014.
Testes for PVT Data from a Laboratory. SPE https://doi.org/10.2118/172396-MS
Reservoir Evaluation & Engineering, v.20, p. 1-16,
2017. https://doi.org/10.2118/183640-PA Taghizadeh, M.; Eftekhari, M. Improved
correlations for prediction of viscosity of Iranian
Pré-Sal Petróleo S.A. Campos Basin - Where It crude oils. Chinese Journal of Chemical
All Starts. Available at: Engineering, v.22, p. 346-354, 2014.
https://www.presalpetroleo.gov.br/ppsa_eng/the- https://doi.org/10.1016/S1004-9541(14)60017-3
pre-salt/campos-basin. Accessed on: 05 July 2019.
Talebi, R.; Ghiasi, M.M.; Talebi, H.;
Rafiee-Taghanaki, S.; Arabloo, M.; Chamkalani, Mohammadyian, M.; Zendehboudi, S.; Arabloo, M.;
A.; Amani, M.; Zargari, M.H.; Adelzadeh, M.R. Bahadori, A. Application of soft computing
Implementation of SVM framework to estimate approaches for modeling saturation pressure of
PVT properties of reservoir oil. Fluid Phase reservoir oils. Journal of Natural Gas Science and
Equilibria, v.346, p. 25-32, 2013. Engineering, v.20, p. 8-15, 2014.
https://doi.org/10.1016/j.fluid.2013.02.012 https://doi.org/10.1016/j.jngse.2014.04.023
151
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
Torabi, F.; Jamaloei, B.Y.; Zunti, C.J.; Markwart, Ostermann and Owolabi (1983)
C. C. The Prediction of Viscosity, Formation Volume
Factor, and Bubble Point Pressure of Heavy Oil R
1.255
T 0.172
55 0.8643 s P0
Using Statistical Analysis, Artificial Neural g API
0.178
Networks, and Three-dimensional Modeling: A
Comparative Evaluation. Energy Sources, v.36, p.
where γg = gas gravity, P = pressure (psi), γAPI = oil
874-889, 2014.
https://doi.org/10.1080/15567036.2010.547926 API gravity, T = temperature (° R).
100.811T 0.497
P
1.2048 10
P0
Rs g 1.4 100.0125 API 0.00091T 460
18.2 X g1.378 Rs1.053100.00069T 4600.0208 API
where γg = gas gravity, P = pressure (psi), γAPI = oil
where γg = gas gravity, P = pressure (psi), γAPI = oil
API (American Petroleum Institute) gravity, T =
temperature (° F). API gravity, T = temperature (° R).
g
1.205
Rs 0.0546 P 2.2 0.00091T 0.0125 API
1.2255
API 2.8869 14.18113.3093log P 0.5
0.989
Rs g 0.172
10 10
10
T 460
where γg = gas gravity, P = pressure (psi), γAPI = oil where γg = gas gravity, P = pressure (psi), γAPI = oil
API gravity, T = temperature (° R). API gravity, T = temperature (° F).
152
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
where γg = gas gravity, P = pressure (psi), γAPI = oil where γg = gas gravity, P = pressure (psi), γo = oil
API gravity, T = temperature (° R). gravity, T = temperature (° R).
X 2
2 API
2.702889206
g 0.5
1.2
44.2500268 0.744335673
P Bo 0.9759 0.00012 Rs 1.25T
o
where γg = gas gravity, P = pressure (psi), γAPI = oil
API gravity, T = temperature (° R). where γg = gas gravity, γo = oil gravity, T =
temperature (° F), Rs = gas-oil ratio (scf/STB).
Hemmati and Kharrat (2007)
1.0857
Glasø (1980)
Rs 0.1769 1.0674
g o5.0956T 0.1394 P
Bo 1 10 X
where γg = gas gravity, P = pressure (psi), γo = oil
gravity, T = temperature (° F). g 0.526
X 6.58511 2.91329log Rs 0.968T 0.27683log R
Mazandarani and Asghari (2007) o
2
g 0.526
T R g
0.526
X 6.58511 o
2.113367 1.4556 5.48944 1.90488
Rs 994.3718 P 2.91329log Rs 0.968T
s 0.968T 0.27683log
g
o o
153
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
where γg = gas gravity, γo = oil gravity, T = where γg = gas gravity, γo = oil gravity, T =
temperature (° F), Rs = gas-oil ratio (scf/STB). temperature (° R), Rs = gas-oil ratio (scf/STB).
API (C C R )
Bo 1 C1 Rs (T 520)
P
2 3 s
Bo 0.408345 1.56667 103 T 1.39775 10 3 X 380525 10 6 X 2 where γo = oil gravity, T = temperature (° F), Rs =
gas-oil ratio (scf/STB).
X Rs0.773572 g0.40402 o0.882605
154
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
X T 0.044
1.52 0.161
o g
g
Khan et al. (1987)
Bo 0.0000009 Rs T 1.0367
o o ob exp 9.6 105 P Pb
where γg = gas gravity, γo = oil gravity, T =
where μob = bubble-point oil viscosity (cP), P =
temperature (° F), Rs = gas-oil ratio (scf/STB).
pressure (psi), Pb = bubble-point pressure (psi).
Torabi et al. (2014)
Abdul-Majeed et al. (1990)
Bo 0.733 0.000014P 0.00842 API 0.00053Rs X 5.21061.11log 6.894757 P Pb
o ob 1000 10
where P = pressure (psi), γAPI = oil API gravity, T =
temperature (° F), Rs = gas-oil ratio (scf/STB). X 1.9311 0.89941ln Rs 0.001194 API
2
0.0092545 API ln R
Rs
Bo 1.1292 0.67343 8.3272 106 T 2 0.00053679Rwhere μob = bubble-point oil viscosity (cP), P =
s 0.0019306T
(T g )2 pressure (psi), Pb = bubble-point pressure (psi), γAPI
8.3272 106 T 2 0.00053679Rs 0.0019306T = oil API gravity, Rs = gas-oil ratio (scf/STB).
155
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
where μob = bubble-point oil viscosity (cP), P = Dindoruk and Christman (2004)
pressure (psi), Pb = bubble-point pressure (psi).
o ob 0.00006334 P Pb 10 X
Labedi (1992)
X C1 C2 log ob C3 log Rs C4 ob log Rs C5 P Pb
102.488 od
0.9036 0.6151
P
ob C3 log Rs C4 ob log Rs C5 P Pb
Pb
o ob X C11 C
2 log
100.01976 API Pb
109.224 where C1 = 0.776644115, C2 = 0.987658646, C3 =
od
4.7013
APIT 0.6739 0.190564677, C4 = 9.147711x10-3, C5 = 1.9111x10-5,
μob = bubble-point oil viscosity (cP), P = pressure
where μob = bubble-point oil viscosity (cP), μod = (psi), Pb = bubble-point pressure (psi), Rs = gas-oil
dead oil viscosity (cP), P = pressure (psi), Pb = ratio (scf/STB).
bubble-point pressure (psi), γAPI = oil API gravity, T =
Temperature (° F). Hossain et al. (2005)
P Pb X10X 1.0146 1.3322 log ob 0.4876 log ob 1.15036 log 7 ln 2 1.48211105 ln 2.27877 104
ob10
2 3
3
b
exp P P 6.5698 ob ob
where μob = bubble-point oil viscosity (cP), P = where μob = bubble-point oil viscosity (cP), P =
pressure (psi), Pb = bubble-point pressure (psi), Rs = pressure (psi), Pb = bubble-point pressure (psi).
gas-oil ratio (scf/STB).
Abedini et al. (2010)
Elsharkawy and Alikhan (1999)
o ob 0.001( P Pb ) X
o ob
10 2.0771
P P
b
1.19279
od
156
BRAZILIAN JOURNAL OF PETROLEUM AND GAS | v. 13 n. 3 | p. 129-157 | 2019 | ISSN 1982-0593
0.3895556 ob
1.226398
o ob P Pb C1 ob
C2 C3 obwhere
0.001254195 Pb 0.3026506
μob = bubble-point oil viscosity (cP), P =
Pb Pb
pressure (psi), Pb = bubble-point pressure (psi), Bob
0.3895556
1.226398
= bubble-point oil formation-volume-factor
C1 ob C2 ob C3 ob 0.001254195 Pb 0.3026506 (bbl/STB), Rs = gas-oil ratio (scf/STB).
Pb Pb
Alqahtani et al. (2018)
where C1 = 7.376096x10-4, C2 = 0.2444663, C3 =
5.60643x10-4, μob = bubble-point oil viscosity (cP), P P
o ob 0.3055 0.1323 X ob 0.0086 0.0099
= pressure (psi), Pb = bubble-point pressure (psi). Pb
P
Taghizadeh and Eftekhari (2014) X 1.0529 0.1317 ob 1.7892 ob 3.4466 0.3579 ob
b
P
o ob exp 8.85 105 ( P Pb )
where μob = bubble-point oil viscosity (cP), P = where μob = bubble-point oil viscosity (cP), P =
pressure (psi), Pb = bubble-point pressure (psi). pressure (psi), Pb = bubble-point pressure (psi).
157