Modeling of Crude Oil Properties Using Artificial Neural Network (ANN)

madbrainedmudlickAI and Robotics

Oct 20, 2013 (3 years and 10 months ago)

144 views


CHEMICAL ENGINEERING



TRANSACTIONS

VOL. 35, 2013

A publication of


The Italian Association

of Chemical Engineering

www.aidic.it/cet

Guest Editors:

Petar Varbanov,
Jiří Klemeš
,

Panos Seferlis, Athanasios

I.

Papadopoulos
, Spyros Voutetakis

Copyright © 2013, AIDIC Servizi S.r.l.,

ISBN

978
-
88
-
95608
-
26
-
6
;
ISSN

1974
-
9791


Modeling of Crude Oil Properties Using Artificial Neural
Network (ANN)

Wirit Cuptasanti
a
,
Farshid Torabi
b
,
Chintana Saiwan
*
a


a

Petroleum and Petrochemical College, Ch
ulalongkorn
University, Bangkok

10330
,
Thailand


b

Petroleum Technology Research Centre, University of Regina, Saskatchewan

S4S7J7, Canada

Chintana
.
Sa
@
chula
.
ac.th

Crude oil properties data were gathered from the published sources for modeling correlations and artifici
al
neural networks (ANN), which could be used to predict crude oil’s physical properties, such as bubble
point pressure and bubble point oil formation volume factor. The data sets were preprocessed for
representativeness of each data points. Each data set
was selected randomly and divided into developing,
and test data sets. The input parameters for bubble point pressure and oil formation volume factor
prediction models were reservoir temperature, solution
-
gas
-
oil ratio, gas specific gravity and API gravity
, or
oil specific gravity. Nonlinear regression was the technique used to develop each correlation. For ANN
development, the developing data sets were randomly divided into training, validation, and testing sets.
Different network architectures and transfe
r functions were used for developing the best ANN models. The
developed correlations and ANNs were tested with the testing data sets, which had not been used for
developing correlations and ANNs, to ensure their accuracy and applicability. Moreover, the de
veloped
models were compared and evaluated with other published correlations. The results showed that the
developed models gave better accuracy in term
s

of average absolute error
s

over other published
correlations.

1.

Introduction

Physical properties of
reservoir fluid are necessary for various field applications, such as field
development, production optimization, reservoir performance evaluation, wellbore hydraulic calculations,
enhanced oil recovery processes,

and

etc. These properties are
essentially
determined at reservoir
temperature with various pressures for reservoir system studies as well as at both various parameters for
wellbore calculations.
It implies that
the fluid properties have
a profound

influence
on

petroleum and
rese
rvoir engineering c
alculations

(
Cosentino, 2001
)
.

T
his valuable information of reservoir fluid
can be

obtained through vast labora
tory testing either on bottom
-
hole or sub
-
surface samples, PVT laboratory
analysis and field production data. In order to acquire the accurate result
s

from the
laboratory, the
reservoir fluid

samples must be appropriately collected and kept for
the
represe
ntative

quality

of the
original reservoir condition. Extremely care and precaution must be taken from
a
specialized sampler since
it is possible to have accurate laboratory results from poor samples,
which leads

to severely fallacious
unrepresentative data result
ing

in devastating consequences throughout the life of a reservoir

(
Cosentino,
2001
)
. Moreover, apart from the
inherent
difficulties of the sampling measurements, these extensive
procedures are very expensive and time
-
consuming.

Since the last century, many empirical correlations of
reservoir fluid properties have

been developed and used in order to solve the aforementioned obstacles.
Therefore, they can be incorporate
d

with reservoir simulations or used in
many
cases where laboratory
data become unavailable. These correlations can predict fluid properties based on

the measured data from
various sources
, but
fail to predict fluid properties in a wide range of conditions
due to

the complexity of
fluid composition of hydrocarbon molecules
,
differen
t

crude characteristic
s

in each area or region
, and
insufficient inform
ation

(
Sutton and Farshad, 1990
)
.
Artificial neural network (ANN) can be another
approach among artificial intelligence techniques for prediction of reservoir fluid physical properties.

The objective of this work wa
s to develop ANN models

a
nd correlations

for predicting

crude oil properties
such as

bubble point pressure and

bubble point

oil formation volume factor using data points available in
literature. The gathered data sets
were

screened, and used for training ANN models. Different neural
model architectures
were

investigated and
then
the best neural model
would be

chosen. Finally, the
developed ANN models
were

evaluated and compared
to

the developed correlations and other

publis
hed

empirical correlations
, using data sets for testing, which had not been used in developing the models
.

2.

M
ethodology

2.1

Data Preparation

It is essential to select the effective inputs to develop an ANN model. However, availability of data is
another major f
actor for choosing the input parameters since ANN demands large volume of data to be
used for training and cross
-
validation in order to solve complex, nonlinear problems accurately. In this
study, large PVT data
were

collected from available
sources and
li
terature.
Collected
Data were checked
based on following criteria.

Redundant data point
s

or

d
ata points with error
s
were
removed.
The data sets
were divided into data sets for developing and

testing ANNs and correlations.

2.2

Developing ANNs and correlations

The prepared data sets were
utilized

for developing ANNs
by
using neural network
toolbox (
nntool)
embedded in Matlab software.
In addition
, the data sets were used for developing correlations

by
using
nonlinear regression technique from Minitab software.
T
he best correlations and ANNs

resulted from
numerous trials were selected.

2.3

Statistical Analysis

As a consequence of testing the developed ANNs and correlations, the statistical parameters, such as
minimum error (
Er
min
), maximum error (
Er
max
), average absol
ute error (
AEr
avg
), and coefficient of
determination (
R
2
) were determined and compared with
the
results from some published correlations.

3.

Results and discussion


3.1

Data Available

For
bubble point pressure

model
ing,
after removing redundant data, a total of 7
57 data points with 3,785
measurements were used.
T
he data
were collected from

Glaso (1980
)
,
Ostermann and Owolabi (1983
)
,
Al
-
Marhoun (1988
)
,
Dokla and Osman (1991
)
,
Omar and Todd (1993
)
,
De Ghetto and Villa (1994
)
,
Mahmood and Al
-
Marhoun (1996
)
,
Velarde et al. (1997
)
,
Gharbi and Elsharkawy (1999
)
,
Wu and
Rosenegger (1999
)
, and
Bello et al. (2008
)
.
The crude oil data consist of reservoir temperature

(
T
res
,
ºF
)
,
solution gas oil ratio

(
R
s
,

scf/
stb
)
, gas specific gravity

(
γ
g
)
, oil API gravity

(
APIº
),

and bubble point
pressure

(
P
b
,

psia
)
.
The da
ta
were randomly
classified
into two sets. A set of 557 data points were used in
developing correlation and ANN, and another set of 200 data points were used for testing the models.
The
data summaries

for developing and testing bubble point pressure models

ar
e shown in T
ables 1
-
2.

Similarly, f
or

the

bubble point oil formation volume f
actor
(B
ob
)
modeling, after removing

redundant data, a
total of 1,175 data points with 5,875 measurements were

used. The data were

collected from
Glaso
(1980
)
,
Ostermann and Owolabi (1983
)
,
Al
-
Marhoun (1988
)
,
Abdul
-
Majeed et al. (1988
)
,
Dokla and Osman
(1991
)
,
Omar and Todd (1993
),
De Ghetto and Villa (199
4
)
,
Mahmood and Al
-
Marhoun (1996
)
,
Gharbi
and Elsharka
wy (1997
)
,
Velarde et

al. (1997
)
,
Wu and Rosenegger (1999
)
, and
Bello et al. (2008
)
.
The
crude
oil data consist of
T
res
,
R
s
,
γ
g
,
API,
and
B
ob
.

The data were randomly divided into two sets
. A set of
875 data points were used in developing correlation and ANN, and another set of 300 data points were
used for testing the models.
The data summar
ies for
developing and testing bubble point oil formation
volume factor are shown in Tables 3
-
4.

Table 1:
Data summary for developing P
b

model
s

(557 points)

Properties

Min

Max

AVG

S.D.

Skewness

Kurtosis

R
s

(scf/
stb
)

8.61

3298.66

639.685

514.78

1.52654

2.95075

T
res

(ºF)

74

341.6

197.875

52.5279

-
0.2074

-
0.5608

γ
g

0.61

3.4445

1.12292

0.42212

1.59279

2.86265

APIº

6

56.8

34.8083

8.27712

-
0.9992

1.44669

P
b

(psia)

79

7127

1978.68

1400.58

0.84701

0.42249


Table 2
:
Data summary for testing P
b

model
s

(200 points)

Properties

Min

Max

AVG

S.D.

Skewness

Kurtosis

R
s

(scf/
stb
)

17.21

3020

657.41

528.524

1.43495

2.55253

T
res

(ºF)

80

334.4

204.357

51.8959

-
0.2426

-
0.1802

γ
g

0.61

2.98

1.16574

0.44579

1.63334

2.9683

APIº

6.3

56.5

35.972

8.41313

-
1.2479

2.17911

P
b

(psia)

95

6641

1970.43

1438.43

0.72348

-
0.0341

Table 3
:
Data summary for developing B
ob

model
s

(875

points)

Properties

Min

Max

AVG

S.D.

Skewness

Kurtosis

R
s

(scf/stb
)

0

3298.66

523.534

480.242

1.66484

3.57134

T
res

(ºF)

74

593.996

187.693

54.1197

0.47773

2.71979

γ
g

0.511

3.4445

1.01727

0.37987

2.08889

5.32879

APIº

6

59.5

32.8496

10.0429

-
0.5984

-
0.341

B
ob

1.028

2.916

1.34781

0.28297

1.77547

4.47046

Table 4
:
Data summary for testing B
ob

model
s

(3
00 points)

Properties

Min

Max

AVG

S.D.

Skewness

Kurtosis

R
s

(scf/stb
)

0

3020

552.867

481.115

1.75751

4.27439

T
res

(ºF)

75.002

341.6

187.153

54.102

0.08627

-
0.6121

γ
g

0.525

2.98

1.03774

0.3922

2.02137

4.72544

APIº

6.3

56.8

33.2908

9.72234

-
0.6421

0.19895

B
ob

1.028

2.903

1.36313

0.29277

2.15482

6.97319


3.2

Developed correlations

After numerous trails on nonlinear regression tec
hnique in Minitab software

using the developing
data
sets
,

the P
b

correlation
was developed

by

modifying

Calhoun
’s correlation

(
Calhoun, 1976
)
, as expressed
in Eq
(
1
)
.


(1)

where
P
b

is a function of
solution gas oil ratio

(
R
s
)
,
reservoir temperature

(
T
res
)
,
oil API gravity

(
API
)
, and
gas specific gravity

(
γ
g
).

Likewise
,
the B
ob

expression in Eq
(
2
)

was correlated using

Petrosky Jr. and Farsahd
’s correlation
(
Petrosky Jr. and Farshad, 1993
)
.


(2
)

where
B
ob

is a function of
solution gas oil ratio (R
s
)
,
reservoir temperature (T
res
)
,
gas specific gravity (
γ
g
),
and oil specific gravity (
γ
o
).


3.3

Developed ANNs

The
similar

developing
data sets as
used for developing

correlations were also used for developing

ANN
models. In this work, 70% of the developing data were randomly used for training, and 30% w
ere

used for
validation and testing each network. Feed
-
forward, back
-
propagation neural network model with one
hidden
-
layer was used for each model. Gradient descent with momentum (GDM) training algorithm and
Levenberg
-
Marquardt (LM) learning algorithm were used for developing the models. Hyperbolic tangent
sigmoid transfer function (TANSIG) was used for calculation betw
een input layer and hidden layer, while
linear transfer function (PURELIN) was chosen
to

calculate

the

output from the hidden layer to the output
layer.

For

P
b

neural network model, four input parameters

including
R
s
, T
res
,

γ
g
,

and API

were

used.
A

neural network with 10

neurons in the hidden layer was
regarded as

the best model

with
the
mean square
error

(MSE)

for validation performance of 1177883.51
. In other words, the 4
-
10
-
1
(input

layer
-
hidden layer
-
output

layer
)
neural network architecture

was

selected
.

For
B
ob

neural network model,
four

input
parameters, which are
R
s
, T
res
,

γ
g
, and
γ
o
,
were used for
B
ob

prediction.
T
he 4
-
1
2
-
1 neural network
architecture was
chosen

for
B
ob

prediction.

A value of the MSE for validation performance of B
ob

ANN is
0.0039261.
The neural network architecture
s

and the
regression plots resulted from the developed
network outputs with respect to targets for training, validation, and testing the developing data sets for

P
b

ANN and B
ob

ANN
are shown in Figure 1
.



a)

b)

Figure 1:

N
eural network

architecture
s

and regression plots resulted from the developed
ANN
s:

a) P
b

ANN,

b) B
ob

ANN
.


3.4

Testing Results

The developed correlations and ANNs were tested against published correlations using data sets for
testing. The testing re
sults from P
b

and B
ob

predictions are shown in Tables 5
-
6.

Table 5
:

Statistical results of P
b

using testing data

Method

Er
min

Er
max

AEr
avg
(%)

AEr
max

(%)

R
2

Standing (1947
)

-
3139.93

1579.84

25.69

372.01

0
.88929

Calhoun (1976
)

-
1882.76

1654.84

53.76

614.8
0

0
.86888

Glaso (1980
)

-
4181.29

1228.53

27.62

24
7.
0
0

0
.87955

Vazquez and Beggs (1980
)

-
3869.91

1307.29

30.15

403.90

0
.8892
0

Al
-
Marhoun (1988
)

-
4049.08

1894.29

23.20

131.6
2

0
.83649

Petrosky Jr. and Farshad (1993
)

-
3035.26

1521.86

86.39

766.86

0
.90579

Dokla and Osman (1991
)

-
1830.1
0

2243.37

29.80

206.23

0
.79883

Kartoatmodjo and Schmidt (1991
)

-
4685.3
0

1179.
75

34.54

487.43

0
.87637

De Ghetto and Villa (1
994
)

-
2617
.00

1624.93

30.22

466.61

0
.89587

Frashad et al. (1996
)

-
1620.52

1624.93

39.08

230.7
7

0
.88458

Almehaideb (1997
)

-
3979.12

1724.54

34.32

427.18

0
.82125

Velarde et al. (1997
)

-
1611.32

2117.66

21.12

110.45

0
.87761

Hanafy et al. (1997
)

-
1882.76

1645.84

53.77

614.8
0

0
.86888

Al
-
Shammasi (1999
)

-
1862.15

1642.29

18.09

105.65

0
.89788

Valkó and McCain Jr (2003
)

-
1566.93

1829.55

18.76

112.73

0
.91467

Dindoruk and Christman (2004
)

-
1314.32

2703.91

25.94

152.31

0
.80465

Nikpoor and Khanamiri (2011
)

-
2731.29

2077.11

20.72

115.4
0

0
.85476

P
b

correlation
(this work)

-
1633.87

1693.7
6

22.36

185.9
2

0
.91846

P
b

ANN (this work)

-
1519.51

1512.67

21.3
2

240
.
19

0
.93176

3.5

Testing Results

The developed correlations and ANNs were tested against published correlations using data sets for
testing. The testing results from P
b

and B
ob

predictions are shown in Tables 5
-
6.

Table 6
:
Statistical results of B
ob

using testing data

Method

Er
min

Er
max

AEr
avg
(%)

AEr
max

(%)

R
2

Standing (1947
)

-
0
.0214

1.594
4

16.70

54.92

0
.81238

Glaso (1980
)

-
0
.167
4

0
.2695

2.84

11.6
1

0
.97351

Al
-
Marhoun (1988
)

-
0
.101
4

0
.2821

1.99

10.
9
0

0
.98026

Al
-
Marhoun (1992
)

-
0
.072
6

0
.577
3

3.5
6

20
.00

0
.97846

Omar and To
dd (1993
)

-
0
.0015

1.6115

17.8
7

55.51

0
.84345

Petrosky Jr. and Farshad (1993
)

-
0
.2337

0
.1530

2.46

15.0
8

0
.97582

Almehaideb
(1997
)

-
0
.206
2

0
.3171

4.2
3

17.73

0
.93238

Hanafy et al. (1997
)

-
0
1.268

0
.1512

7.97

43.93

0
.93602

Al
-
Shammasi
(1999
)

-
0
.2136

0
.4123

3.06

16.66

0
.95197

Hemmati and Kharrat (2007
)

-
0
.178
9

0
.1805

1.8
9

11.53

0
.98179

Nikpoor and Khanamiri (2011
)

-
0
.1421

0
.412
9

2
.00

14.30

0
.97513

B
ob

correlation
(this work)

-
0
.137
7

0
.2189

1.67

8.2
1

0
.9839
5

B
ob

ANN (this work)

-
0
.195
1

0
.187
6

2.13

9.6
7

0
.9
8134


Regarding

the

P
b

predic
tion

results
, the developed P
b

ANN
gave competitive performance compared to
some of the correlations.
Although

the
A
Er
avg

(21.32

%)
from the developed P
b

ANN was higher than some

published

correlations
,

the developed P
b

ANN

had the highest R
2

of 0.93176
with t
he narrowest range
between
Er
min

(
-
1519.51)
and Er
max

(
1512.67)
.

T
he developed
P
b

correlation (
Equation 1
)

also gave
competitive performance

for P
b

prediction

compared to
most of the published correlations

with R
2

of
0.91846.
For the prediction of B
ob
,
both

developed

B
ob

correlation

(Equation 2)

and
B
ob

ANN outperformed
the published correlations in term of AEr
max
, which are 8.21

% for B
ob

correlation and 9.67

% for B
ob

ANN
.
Moreover
,

the

result f
r
om the

developed B
ob

correlation

was

slightly

better than B
ob

ANN

in term of R
2

(0.98134)
and
A
Er
avg

(1.67

%)
.

4.

Conclusions


The ANNs and correlations
were developed
for
the
prediction of
bubble point pressure

and
bubble point o
il
formation volume factor

using data gathered from
various
published sources. The prepared data sets were
divided into data sets for developing and testing correlations and ANNs. The developed
P
b

ANN and the
P
b

correlation
could competitively

predict

bubble point pressure

when compared

to the other published
corre
lations.
On the other hand
, the
developed

B
ob

ANN

and

B
ob

correlation
could be
satisfactorily
employed in the

prediction of
bubble point o
il formation volume factor

under an acceptable
range of data
.

Acknowledgements

I would like to thank University of
Regina, Petroleum Technology Research Centr
e for research funding.
Also
thanks to Petroleum and Petrochemical College and Petroleum and Petrochemical Research Unit
Center for Petroleum, Petrochemicals, and Advanced Materials, Chulalongkorn University, Thai
land.


References

Abdul
-
Majeed

G.H.
, Salman N.H., Scarth

B.R., 1988, An
e
mpirical
c
orrelation for
o
il Fvf
p
rediction, Journal
of Canadian Petroleum Technology, 27.

Al
-
Marhoun

M.A., 1988, PVT Correlations for
M
iddle
E
ast
c
rude
o
ils, SPE Jo
urnal of Petroleum
Technology, 40, 650
-
666.

Al
-
Mar
houn

M.A., 1992, New
c
orrelations for

f
ormation
v
olume
f
actors of
o
il and
g
as
m
ixtures, Journal of
Canadian Petroleum Technology, 31.

Al
-
Shammasi

A.A., 1999, Bubble Point
p
ressure and
o
il
f
ormation
v
olume
f
actor
c
orrelations, Middle East
Oil Show and Conference. Bahrain: Copyright 1999, Society of Petroleum Engineers Inc.

Almehaideb

R.A., 1997, Improved PVT Correlations for UAE Crude Oils, Middle East Oil Show and
Conference. Bahrain: 1997 Copyright 1997, Society of Petroleum Engineers, Inc.

Bello

O.O., Reinicke K.M., Patil P.A., 2008, Comparison of the
p
erformance of
e
mpirical
m
odel
s
u
sed for
the
p
rediction of the PVT
p
roperties of
c
rude
o
ils of the Niger Delta, Petroleum Science and
Technology, 26, 593
-
609.

Calhoun

J.C.J., 1976, Fundamentals of reservoir engineering.

Cosentino

L., 2001, Integrated Reservoir Studies, Editions Technip
.

De Ghetto G., Villa

M., 1994, Reliability
a
nalysis on PVT
c
orrelations, European Petroleum Conference.
London, United Kingdom: 1994 Copyright 1994, Society of Petroleum Engineers, Inc.

Dindoruk B., Christman

P.G., 2004, PVT
p
roperties and
v
iscosity
c
orre
lations for Gulf of Mexico
o
ils, SPE
Reservoir Evaluation & Engineering, 7, 427
-
437.

Dokla

M.E., Osman M.E., 1991, Correlation of PVT
p
roperties for U.A.E. Crudes, Middle East Oil Show

and Conference
. Bahrain: 1990.

Frashad F., Leblanc J.L., Garber

J.D., O
s
orio J.G., 1996, Empirical PVT c
orrelations for Colombian
c
rude
o
ils.

Gharbi R.B., Elsharkawy

A.M., 1997, Universal
n
eural
n
etwork
b
ased
m
odel for
e
stimating
t
he PVT
p
roperties of
c
rude
o
il
s
ystems, SPE Asia Pacific Oil and Gas Conference and Exhibition.
Kuala
Lumpur, Malaysia: 1997 Copyright 1997, Society of Petroleum Engineers, Inc.

Gharbi R.B.C., Elsharkawy

A.M., 1999, Neural
n
etwork
m
odel for
e
stimating the PVT
p
roperties of Middle
East
c
rude
o
ils, SPE Reservoir Evaluation & Engineering, 2, 255
-
265.

Glaso

O., 1980, Generalized
p
ressure
-
v
olume
-
t
emperature
c
orrelations, SPE Journal of Petroleum
Technology, 32, 785
-
795.

Hanafy H.H., Macary S.M., Elnady Y.M., Bayomi A.A., El Batanony M.H., 1997, A
n
ew
approach for
p
redicting the
c
rude
o
il
p
roperties, SPE
Production Operations Symposium. Oklahoma City, Oklahoma:
1997.

Hemmati

M.N., Kharrat R., 2007, A Correlation
a
pproach for
p
rediction of
c
rude
-
o
il PVT
p
roperties, SPE
Middle East Oil and Gas Show and Conference. Bahrain: Society of Petroleum Engineers
, Inc
.

Kartoatmodjo

T.R.S., Schmidt Z., 1991, New
c
orrelations for
c
rude
o
il
physical p
roperties. Society of
Petroleum Engineers
, Inc
.

Mahmood M.A., Al
-
Marhoun M.A., 1996, Evaluation of empirically derived PVT properties for Pakistani
crude oils, Journal of Pet
roleum Science and Engineering, 16, 275
-
290.

Nikpoor M.H., Khanamiri

H.H., 2011, New
e
mpirical
c
orrelations for
p
redicting
c
rude
o
il
p
roperties,
Petroleum Science and Technology, 29, 1649
-
1658.

Omar

M.I., Todd A.C., 1993, Development of
n
ew
m
odified
b
lack
o
il
c
orrelations for Malaysian
c
rudes,
SPE Asia Pacific Oil and Gas Conference. Singapore: 1993 Copyright 1993, Society of Petroleum
Engineers, Inc.

Ostermann

R.D., Owolabi O.O., 1983, Correlations for the
r
eservoir
f
luid
p
roperties of Alaskan
c
rudes,
SPE
California Regional Meeting. Ventura, California: 1983 Copyright 1983 Society of Petroleum
Engineers of AIME.

Petrosky Jr. G.E., Farshad

F.F., 1
993, PVT

c
orrelations for Gulf of Mexico
c
rude
o
ils, SPE Annual
Technical Conference and Exhibition. Houston, Te
xas: 1993 Copyright 1993, Society of Petroleum
Engineers, Inc.

Standing

M.B., 1947, A P
VT

c
orrelation
f
or
m
ixtures
o
f California
o
ils
and g
ases. American Petroleum
Institute.

Sutton R.P., Farshad

F., 1990, Evaluation of
e
mpirically
d
erived PVT
p
roperties f
or Gulf of Mexico
c
rude
o
ils, SPE Reservoir Engineering, 5, 79
-
86.

Valkó P.P., Mccain Jr

W.D., 2003, Reservoir oil bubblepoint pressures revisited; solution gas

oil ratios and
surface gas specific gravities, Journal of Petroleum Science and Engineering, 37
, 153
-
169.

Vazquez M., Beggs

H.D., 1980, Correlations for
fluid p
hysical
property p
rediction, SPE Journal of
Petroleum Technology, 32, 968
-
970.

Velarde J., Blasingame T.A., W. D. Mccain

J., 1997, Correlation of
b
lack
o
il
p
roperties at
p
ressures
below
b
ubble
p
oint
p
ressure
-

A New a
pproach,
a
nnual
t
echnical
m
eeting. Calgary, Alberta: Petroleum
Society of Canada.

Wu R., Rosenegger

L., 1999, Comparison of PVT
p
roperties from
e
quation of
s
tate
a
nalysis and PVT
c
orrelations for
r
eservoir
s
tudies, Annual
t
ech
nical
m
eeting. Calgary, Alberta: Petroleum Society of
Canada.