Performanceof

Proceedings of the 2022 14th International Pipeline Conference
IPC2022
September 26-30, 2022, Calgary, Alberta, Canada
IPC2022-86297
Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

PERFORMANCE OF TWELVE DIFFERENT EQUATIONS OF STATE FOR NATURAL GAS
AND HYDROGEN BLENDS
Kamal K. Botros Larry Jensen

NOVA Chemicals Corporation TC Energy
Centre for Applied Research Calgary, Alberta, Canada
Calgary, Alberta, Canada larry_jensen@tcenergy.com
kamal.botros@novachem.com
ABSTRACT performing EOS for all ranges of P and T, while the second-
The aspiration for blending hydrogen (H2) into natural gas best performer is BWRS in the range of 0.1-3 MPa and AGA8
(NG) in gas transmission systems is high and is happening in the range of 3-25 MPa, respectively.
globally. However, the principal properties of the blended
mixtures and their thermodynamic derivatives can
significantly vary depending on the Equation of State (EOS) NOMENCLATURE
employed. There is a need to arrive at the best performing EOS
ATR Auto-Thermal Reformer
for the prediction of the blended mixtures from low to high
C Speed of sound in the gas mixture (m/s)
concentration of H2 in the blend with NG. Twelve different
Cp Isobaric heat capacity (J/kg. oC)
EOS were evaluated against measured data found in the open
E Error in property predictions defined by Eq. (3)
literature of pure H2, binary mixtures with alkanes and
EOS Equation of State
mixtures with NG. Three measured properties were found,
h Enthalpy (J/kg)
namely density, speed of sound and isobaric heat capacity
J.T. Joule-Thomson coefficient (oC/MPa)
(Cp) in the range of pressures up to 50 MPa and temperatures
MW Gas molecular weight (kg/kmole)
in the range of -20oC to +80oC and H2 concentration up to
n Number of data points for a given set
88%(mole). The total number of measured data points are 629
NG Natural Gas
for pure H2 and 1788 for Binaries and NG mixtures with H2.
P Pressure (MPa)
Performance of each EOS is based on the average of the
s Entropy (J/kg. oC)
absolute error (deviation%) between predicted vs. measured
SMR Steam Methane Reformer
parameters. These were: density, which represents the
T Temperature (oC)
principal performance of the EOS with respect the basic
W.T. Pipe wall thickness (mm)
formulation of P,  and T, the speed of sound, which represents
Ym Measured thermodynamic property (, C or Cp)
a thermodynamic derivative with respect to entropy, and
isobaric heat capacity which represents a thermodynamic Yp Predicted thermodynamic property (, C or Cp)
derivative with respect to enthalpy. All other thermodynamic
derivatives can be related to these three parameters (e.g., J.T. Greek Letters:
coefficient, isochoric heat capacity, enthalpy, internal energy,  Density (kg/m3)
, compressibility factor and Helmholtz and Gibbs free
energies, etc.). Transport properties predictions are based on
other empirical and semi-empirical correlations that are 1 INTRODUCTION
independent of the EOS and hence were not considered. It was Gas pipeline companies are actively pursuing extending
found that, for the most part and for pure H2, GERG2008 EOS their business opportunities to transport clean energy
performed best in predicting the above three principal hydrogen either from renewable sources or from valorizations
parameters followed by AGA8. For Binary and NG mixtures processes (e.g. SMR, ATR, etc.), as part of their commitment
with H2, again GERG2008 was found to be the best to climate change and GHG reduction initiatives. One mode of
1 Copyright © 2022 by ASME

transportation of hydrogen from source to market is by compressibility factor, J.T. coefficient, etc.). Additionally, the
blending with natural gas (NG) flowing through existing prediction of transport properties is based on other empirical
pipeline transmission networks (pressure in the range of 3-10 and semi-empirical correlations that are independent of the
MPa) and distribution systems (0.1-3 MPa), or via dedicated EOS and hence were not considered.
pipelines exclusively designed and built to transport pure H2. The 12 EOS considered in the present work are:

The first mode requires employment of blending stations at H2 1. GERG2008 [1].
source locations capable of effectively mixing a wide range of 2. Cubic-Plus-Association (CPA) [2,3].
H2 concentrations with natural gas streams flowing through 3. AGA8 [4-6].
existing pipeline systems. Compression facilities need to be 4. Benedict-Webb-Rubin-Starling (BWRS) [8].
configured and modified if necessary to accommodate 5. Benedict-Webb-Rubin-Lee-Starling (BWR-LS)
variations in the blend compositions particular the mole% of [9,10].
H2 in the blend. Operations of these systems are generally 6. Peng-Robinson (PR) [11].
unsteady in that they are subject to packing-drafting, receipt 7. PR-with mixing rules by Huron and Vidal
vs. delivery capacity changes, as well as interruptions in the (implemented in Aspen as PR-MHV2) [12].
operations of compression facilities due to scheduled or 8. PR- with mixing rules by Boston and Mathias (PR-
unscheduled maintenance and repairs. BM) [13].
Network overall operations, whether steady or unsteady, 9. PR-with mixing rule by Wong and Sandler (PR-
are monitored live and the hydraulics analyzed continuously. WS) [14].
The accuracy of such analyses requires high fidelity models 10. Soave- Redlich-Kwang (SRK) [15,16].
(steady or transient) and incorporation of an appropriate 11. RKS- modified by Boston and Mathias (RKS-BM)
equation of state (EOS) to predict necessary properties to [13].
perform various simulations. Of utmost importance are the 12. RKS- with mixing rule by Wong and Sandler
relationships between mixture density (compressibility), speed (RKS-WS) [14].
of sound and enthalpy with pressures, temperature and
composition of the gas mixture blends. The above 12 EOS will be referred to, hence force, by the
The present paper aims at providing performance designation in bold fonts between brackets as indicated above.
assessment of a total of 12 EOS commonly used in the gas
industry specifically with respect to (w.r.t.) their individual
predictions of the aforementioned three principal parameters, 2 APPROACH AND METHODOLOGY
i.e. density, speed of sound and the isobaric heat capacity (Cp). Extensive literature search was conducted to collect all
Density is very critical parameter in hydraulic simulations and published measured data of the three principal thermodynamic
represents the performance of the EOS w.r.t. P,  and T properties mentioned above, namely density, speed of sound
correlation. The speed of sound is critical as it, together with and Cp. A total of 629 measured data points were found for
density, represents the impedance of travelling perturbations pure (100%) hydrogen, 386 of which are measured density,
along the pipeline system. It also represents the performance 159 are measured speed of sound and 84 are measured Cp [17-
of the respective EOS w.r.t. entropy, s, since the speed of 21]. The pressure (P) ranges from atmospheric to 50 MPa,
sound is defined as: while temperature (T) ranges from -75oC to +75oC [22-61].
These data are plotted in Figs. 1 through 3, for density, speed
 P  (1) of sound and Cp shown on the respective x-axis, while the
C=   corresponding P and T are shown on two independent y-axes
   s
on the left and right, respectively.
Similarly, an extensive literature search was conducted to
Lastly, the isobaric heat capacity (Cp) defined as: collect published measured data of the same three principal
thermodynamic properties for natural gas or binary mixtures
 h  (2)
Cp =  
with hydrogen. A total of 1788 measured data points were
 T  p found for Binaries and NG mixtures with hydrogen [22-61],
1506 of which are measured density, 282 are measured speed
represents the performance of the respective EOS w.r.t. of sound and no measurements were found for Cp. The
enthalpy (h) which plays an important role in heat transfer pressure ranges also from atmospheric to 50 MPa, while
simulations, e.g. the temperature profiles/variations along a temperature ranges from -100oC to +200oC. The concentration
pipeline, enthalpy rise (head) across a compressor unit, heat of H2 varied from 3% (mole) to 88.5% (mole). The other
transfer through aerial gas coolers, Joule-Thomson effects components are all alkanes (up to hexane), helium, nitrogen,
across a throttling element, etc. carbon dioxide, oxygen and water vapour. Figure 4 shows
representative plots of the composition for the top 6 most
All other thermodynamic derivatives can be related to the components according to their respective concentrations. The
above three principal parameters (e.g., isochoric heat capacity, corresponding ‘zero’ values shown along the x-axis mixtures
internal energy, Helmholtz and Gibbs free energy, that have the indicated hydrogen more fraction but do not have

that indicated labeled component on the respective y-axis in
them.
These data are also plotted in Figs. 5 and 6, for density and
speed of sound on the respective x-axis, while the
corresponding P and T are again shown on two independent y-

axes on the left and right, respectively.
Figure 2: Data of Measured Speed of Sound for Pure

Hydrogen at Different Pressures and Temperatures.
Figure 1: Data of Measured Densities for Pure Hydrogen

at Different Pressures and Temperatures.
Figure 3: Data of Measured Isobaric Heat Capacity for

Pure Hydrogen at Different Pressures and
Temperatures.

Figure 4: Composition of Various Binaries and NG mixtures with Hydrogen.
Figure 5: Data of Measured Densities for Binary and NG Figure 6: Data of Measured Speed of Sound for Binary
Mixtures with Hydrogen at Different Pressures and and NG Mixtures with Hydrogen at Different Pressures
Temperatures. and Temperatures.

The first two rows in Table 1 (in the Appendix) summarize of density is 0.033%, and in the speed of sound is 0.213%,
the data points for both pure (100%) H2 and for Binaries and while that for Cp is 0.151%. The sum of these three averages
NG mixtures with H2, respectively. Performance evaluations of the absolute errors, E, is 0.397%, which here in after is also
are first conducted on these two sets of data separately (see taken as a fourth principal criterion in the performance
Section 3 and Section 4, respectively). Three subsets were then evaluation of all EOS considered. Here the overall error, E, is

extracted from the Binaries and NG mixtures in three ranges defined as:
of pressures and temperatures as follow:
1. Subset #1 is for the range of P = 0.1-3 MPa, and range of E (%) = E + EC + ECp (4)
T = -20oC to +80oC, as these ranges are most applicable avg avg avg
to low pressure distribution systems. Results are

presented in Section 5. where the subscripts “”, “C” and “Cp” indicate density, speed
2. Subset #2 is for the range of P = 3-10 MPa, and range of of sound and isobaric heat capacity, respectively. The same
T = -20oC to +80oC, as these ranges are most applicable procedure is applied to all other 11 EOS, and the results in
to gas transmission networks from receipt to high- terms of the respective average of the absolute errors in the
pressure industrial delivery points and city gates. Results predictions of the three thermodynamic parameters as well as
are presented in Section 6. the combined sum of these errors (E) are tabulated in Table 2.
3. Subset #3 is for the range of P = 10-25 MPa, and range The respective EOS as implemented in ASPEN Plus (version
of T = -20oC to +80oC, as these ranges are most 11.0) are used henceforth in the present work. It was found
applicable to production and blending facilities upstream that RKS-WS EOS did not perform well for all data points and
of the pipeline transmission systems. Results are all three thermodynamic properties for pure hydrogen, while
presented in Section 7. PR-WS EOS, also did not perform well for density and speed
The bottom three rows in Table 1 provides details about of sound predictions.
each of the above three subsets and the number of measured Based on the results in Table 2, all other 10 EOS (i.e.
data points available for the analysis. discounting RKS-WS and PR-WS) were ranked according to
The performance criterion is based on taking the average of their respective predictions of the three thermodynamic
the absolute error (%) among applicable measured data points parameters as well as the overall error parameter, E, in Fig. 8.
for each principal property, defined as: It is clearly shown that GERG2008 outperforms all other EOS
in all categories. Other observations can be listed below:
1 Yp − Ym 1. Second best performing EOS in the prediction of density
Avg. of Abs. Error(%) = 
n n Ym
x 100% (3)
is RKS-BM, while SRK is ranked next.
2. Second best performing EOS in the prediction of speed of
sound is also RKS-BM, while PR-MHV2, PR-BM and
where, Y is the thermodynamic property under evaluation, SRK are ranked next.
subscripts ‘p’ and ‘m’ correspond to predicted and measured 3. Second best performing EOS in the prediction of Cp is
values, respectively, and n is the number of data points PR-MHV2 and PR-BM, while CPA, BWR-LS and
available for the evaluation for the specific thermodynamic BWRS are ranked next.
property. It should be noted that for the case where no 4. Second best performing EOS in terms of the overall
measured Cp data are available, errors associated with combined error criterion, E, (bottom chart in Fig. 8) is
GERG2008 was assumed to be comparable to average error in RKS-BM, while SRK is ranked next.
density and speed of sound, while errors in other EOS are
taken relative to the prediction by GERG2008. This will be
elaborated further throughout the results Sections to follow.
3 RESULTS FOR PURE HYDROGEN

We begin by evaluation of the performance of the selected
12 EOS in predicting the density, speed of sound and Cp for
pure hydrogen. Figure 7 illustrates the results of the evaluation
based on GERG2008 EOS for these three thermodynamic
properties, where the respective measured values are given in
Figs. 1 through 3, earlier, along with the corresponding
number of data points available. The plots on Fig. 7 depicts
the actual error including the +/- sign, whereas the imbedded
values are the average of the absolute values as defined in Eq.
(3). All predictions were based on GERG2008 as implemented
in ASPEN Plus (version 11.0), and further verified by
GERG2008 as implemented in REFPRP (version 9.1) [62]. It
is shown that the average of the absolute error in the prediction

Table 2: Summary Performance Table of the 12 EOS in
Terms of Density, Speed of Sound and Heat Capacity
Prediction Errors for Pure Hydrogen.
Pure H2
# Average of Absolute Error (%) Sum of Absolute

E.O.S. Errors (%) (Principal
Density Speed of Sound Heat Capacity (Cp ) Criterion)
1 GERG2008 0.033 0.213 0.151 0.397
2 CPA 2.190 1.490 0.855 4.535
3 AGA8 2.640 2.780 1.040 6.460
4 BWRS 4.630 4.890 0.948 10.468
5 BWR-LS 4.520 4.620 0.931 10.071
6 PR 1.770 1.510 1.550 4.830
7 PR-MHV2 2.610 1.310 0.571 4.491
8 PR-BM 2.610 1.310 0.571 4.491
9 PR-WS Poor Performance 2.620 -
10 SRK 1.220 1.320 0.971 3.511
11 RKS-BM 0.530 0.625 1.020 2.175
12 RKS-WS Poor Performance
Figure 7: Example of Density, Speed of Sound and Heat

Capacity Prediction Errors Based on GERG2008 EOS
for Pure Hydrogen.

4 RESULTS FOR BINARIES AND MIXTURES
OF NG WITH HYDROGEN
Recall from Table 1 (Appendix), that measured data for
binaries and NG mixtures with hydrogen is limited only to

densities and speeds of sound and none were found for Cp.
Similarly, Fig. 9 illustrates the results of the evaluation based
on GERG2008 EOS for these two thermodynamic properties,
where the respective measured values are given in Figs. 4 and
5, earlier, along with the corresponding number of data points
available. Again, the plots on Fig. 9 depicts the actual error
including the +/- sign, whereas the imbedded values are the
average of the absolute values as defined in Eq. (3). It is shown
the average of the absolute error in the prediction of density is
0.084%, and in the speed of sound is 0.039%. Since there is no
measured data available for Cp, an average value of E for
density and speed of sound predictions of 0.062% is assumed
for Cp. The primary reason for doing this is allow comparison
of the other 11 EOS to each other taking GERG2008 as a
reference EOS based on its superior performance for pure
hydrogen and the three thermodynamic properties, and also for
the present binaries and mixtures with H2 for density and
speed of sound exhibited in Fig. 9.
The same procedure is applied to all other 11 EOS, and the
results are given in Table 3. Unlike the case in pure H2, all
EOS produced results as opposed to no results at all. Based on
the results in Table 3, EOS were ranked according to their
respective predictions of the three thermodynamic parameters
as well as their respective overall error parameter, E, in Fig.
10. It is clearly shown that, again GERG2008 outperforms all
other EOS in all categories. Other observations are listed
below:
1. Second best performing EOS in the prediction of density
is SRK, while AGA8, CPA, RKS-BM and PRRKS-BM
are ranked next .
sound is AGA8, while PR-BM, PR-MHV2, PR-WS and
PR are ranked next .
3. Second best performing EOS in the prediction of Cp is
BWRS, while BWR-LS and CPA are ranked next .
4. Second best performing EOS in terms of the overall
combined error criterion, E, (bottom chart in Fig. 10) is
AGA8, while CPA, PR and SRK are ranked next .
Figure 8: Performance Ranking of the 12 EOS in Terms
of Density, Speed of Sound and Heat Capacity
Prediction Errors for Pure Hydrogen.

Figure 9: Example of Density and Speed of Sound
Prediction Errors Based on GERG2008 EOS for
Binaries and NG Mixtures with Hydrogen.
Table 3: Summary Performance Table of the 12 EOS in

Terms of Density, Speed of Sound and Heat Capacity
Prediction Errors for Binaries and NG Mixtures with
Hydrogen.
Binary & NG Mixtures with H2

Density Speed of Sound Heat Capacity (Cp)* Criterion)
1 GERG2008 0.084 0.039 0.062 0.185
2 CPA 0.871 0.933 0.439 2.243
3 AGA8 0.836 0.453 0.546 1.835
4 BWRS 1.700 2.530 0.321 4.551
5 BWR-LS 1.590 3.210 0.376 5.176
6 PR 0.886 0.824 0.687 2.397
7 PR-MHV2 2.090 0.664 1.900 4.654
8 PR-BM 1.530 0.468 0.940 2.938

9 PR-WS 2.140 0.733 3.180 6.053 of Density, Speed of Sound and Heat Capacity
10 SRK 0.480 1.390 0.599 2.469 Prediction Errors for Binaries and NG Mixtures with
11 RKS-BM 0.879 2.690 1.070 4.639 Hydrogen.
12 RKS-WS 2.250 7.290 3.610 13.150
* Since no measured data of Cp are available, error associated with GERG2008 was assumed to be
equal to average error of density and speed of sound, while errors in other EOS are w.r.t.
GERG2008

5 RESULTS FOR BINARIES AND MIXTURES Table 4: Summary Performance of the 12 EOS in Terms
OF NG WITH HYDROGEN IN THE RANGE OF of Density, Speed of Sound and Heat Capacity
0.1-3 MPA Prediction Errors for Binaries and NG Mixtures with
In this Section, a sub-set from the binaries and NG mixtures Hydrogen in the Range of 0.1-3 MPa.

with H2 was extracted in the range of pressure between 0.1
MPa and 3 MPa, and temperatures in the range between -20oC Binary & NG Mixtures with H2 (0.1-3 MPa &-20oC to +80oC)
and +80oC, as this range, particularly the pressure, would be # Average of Absolute Error (%) Sum of Absolute
of interest to pipeline and distribution systems. Since there is E.O.S. Errors (%) (Principal
no measured data available for Cp for these mixtures, an Density Speed of Sound Heat Capacity (Cp)* Criterion)
average value of E between density and speed of sound
1 GERG2008 0.055 0.040 0.048 0.143
predictions for GERG2008 is assumed for Cp and specific to
the above ranges of P and T. For this range, the error E (as 2 CPA 0.120 0.211 0.152 0.483
defined by Eq. 3) in GERG2008 predictions of density was 3 AGA8 0.142 0.086 0.316 0.544
found to be 0.055%, and in the prediction of the speed of sound 4 BWRS 0.082 0.107 0.119 0.308
is 0.040%, as shown in the first row of Table 4. Hence the
5 BWR-LS 0.130 0.106 0.127 0.363
assumed value of E in the prediction of Cp by the same EOS
would be 0.048%. The performance of the other 11 EOS 6 PR 0.177 0.244 0.187 0.608
predictions is referenced to GERG2008 predictions in so far 7 PR-MHV2 0.634 0.139 0.416 1.189
only to Cp. I 8 PR-BM 0.438 0.131 0.252 0.821
Performance summary of all 12 EOS is compiled in the rest
9 PR-WS 2.560 0.187 0.834 3.581
of Table 4 and ranked with respect to the same three respective
three thermodynamic properties, as well as their respective 10 SRK 0.082 0.322 0.168 0.572
overall error parameter, E, in Fig. 11. Similarly, GERG2008 11 RKS-BM 0.116 0.535 0.261 0.912
was found to outperform all other EOS in all categories. The 12 RKS-WS 1.840 0.551 0.904 3.295
second best and ‘accepted’ EOS can be drawn from Fig. 11: * Since no measured data of Cp are available, error associated with GERG2008 was assumed to be
1. Second best performing EOS in the prediction of density equal to average error of density and speed of sound, while errors in other EOS are w.r.t.
are BWRS and SRK, while RKS-BM,CPA,BWR-LS and GERG2008
AGA8 are ranked next .

sound is AGA8, while BWR-LS and , BWRS are ranked
next .
3. Second best performing EOS in the prediction of Cp are
BWRS and BWR-LS, while CPA and SRK are ranked
next .
combined error criterion, E, (bottom chart in Fig. 11) are
BWRS and BWR-LS, while CPA, AGA8 and SRK are
ranked next .

6 RESULTS FOR BINARIES AND MIXTURES
OF NG WITH HYDROGEN IN THE RANGE OF
3-10 MPA
Another sub-set was extracted from the binaries and NG
mixtures with H2 in the range of pressure between 3 MPa and

10 MPa, and temperatures in the range between -20oC and
+80oC, as this range, would be of interest to conventional gas
transmission systems from production facilities (receipt
stations) to delivery points along the system including
industrial setup and city gates. Again, since there is no
measured data available for Cp for these mixtures, an average
value of E between density and speed of sound predictions for
GERG2008 is assumed for Cp and specific to the above ranges
of P and T. In this case, the error E (as defined by Eq. 3) in
GERG2008 predictions of density was found to be 0.128%,
and in the prediction of the speed of sound is 0.036%, as shown
in the first row of Table 5. Hence the assumed value of E in
the prediction of Cp by the same EOS would be 0.082%. Like
in the previous two Sections, the performance of the other 11
EOS predictions is referenced to GERG2008 predictions in so
far only to Cp.
Performance summary of all 12 EOS is compiled in Table 5
and ranked with respect to the same three respective three
thermodynamic properties, as well as their respective overall
error parameter, E, in Fig. 12. Similarly, GERG2008 was
found to outperform all other EOS in all categories. The
second best and ‘accepted’ EOS can be drawn from Fig. 12:
is SRK, while BWR-LS and BWRS are ranked next.
sound are AGA8 and PR-BM, while PR-WS, PR-MHV2
and PR are ranked next.
3. Second best performing EOS in the prediction of Cp is
BWRS, while BWR-LS, CPA, AGA8, SRK and PR are
ranked next.
AGA8, while BWRS, CPA and PR are ranked next.

Hydrogen in the Range of 0.1-3 MPa.

Table 5: Summary Performance of the 12 EOS in Terms
Hydrogen in the Range of 3-10 MPa.

Binary & NG Mixtures with H2 (3-10 MPa &-20oC to +80oC)

Density Speed of Sound Heat Capacity (Cp)* Criterion)
1 GERG2008 0.128 0.036 0.082 0.246
2 CPA 0.574 0.841 0.457 1.872
3 AGA8 0.572 0.332 0.555 1.459
4 BWRS 0.547 0.944 0.251 1.742
5 BWR-LS 0.485 1.390 0.365 2.240
6 PR 0.711 0.625 0.611 1.947
7 PR-MHV2 1.960 0.549 1.830 4.339
8 PR-BM 1.400 0.387 0.951 2.738
9 PR-WS 2.150 0.543 2.950 5.643
10 SRK 0.382 1.330 0.563 2.275
11 RKS-BM 0.694 2.390 1.040 4.124
12 RKS-WS 1.820 3.280 3.770 8.870

* Since no measured data of Cp are available, error associated with GERG2008 was assumed to be
equal to average error of density and speed of sound, while errors in other EOS are w.r.t.
GERG2008

Hydrogen in the Range of 3-10 MPa.

7 RESULTS FOR BINARIES AND MIXTURES Table 6: Summary Performance of the 12 EOS in Terms
OF NG WITH HYDROGEN IN THE RANGE OF of Density, Speed of Sound and Heat Capacity
10-25 MPA Prediction Errors for Binaries and NG Mixtures with
A third sub-set from the binaries and NG mixtures with H2 Hydrogen in the Range of 10-25 MPa.

was extracted in the range of pressure between 10 MPa and 25
MPa, and temperatures in the range between -20oC and +80oC, Binary & NG Mixtures with H2 (10-25 MPa &-20oC to +80oC)
as this range, particularly the pressure, would be of interest to # Average of Absolute Error (%) Sum of Absolute
production and blending facilities upstream of conventional E.O.S. Errors (%) (Principal
NG pipeline transmission, and also in industrial and Density Speed of Sound Heat Capacity (Cp)* Criterion)
petrochemical industries requiring relatively high pressures

1 GERG2008 0.052 0.041 0.047 0.140
for process requirements. Again, since there is no measured
data available for Cp for these mixtures, an average value of E 2 CPA 1.570 1.400 0.531 3.501
between density and speed of sound predictions for 3 AGA8 1.420 0.750 0.639 2.809
GERG2008 is assumed for Cp and specific to the above ranges 4 BWRS 3.350 4.900 0.464 8.714
of P and T. In this case, the error E (as defined by Eq. 3) in
5 BWR-LS 3.130 6.070 0.497 9.697
GERG2008 predictions of density was found to be 0.052%,
and in the prediction of the speed of sound is 0.041%, as shown 6 PR 1.490 1.230 1.030 3.750
in the first row of Table 4. Hence the assumed value of E in 7 PR-MHV2 2.930 1.060 2.830 6.820
the prediction of Cp by the same EOS would be 0.047%. 8 PR-BM 2.240 0.741 1.450 4.431
Again, the performance of the other 11 EOS predictions is
9 PR-WS 2.330 1.200 2.790 6.320
referenced to GERG2008 predictions in so far only to Cp.
10 SRK 0.782 1.980 0.838 3.600
Performance summary of all 12 EOS is compiled in Table 6
and ranked with respect to the same three respective three 11 RKS-BM 1.680 4.000 1.640 7.320
thermodynamic properties, as well as their respective overall 12 RKS-WS 3.550 6.500 6.950 17.000
error parameter, E, in Fig. 13. Again, GERG2008 outperforms * Since no measured data of Cp are available, error associated with GERG2008 was assumed to be
all other EOS in all categories. The second best and ‘accepted’ equal to average error of density and speed of sound, while errors in other EOS are w.r.t.
EOS can be drawn from Fig. 13: GERG2008

is SRK, while AGA8 and PR are ranked next .
sound are PR-BM and AGA8, while PR-MHV2, PR-WS
and PR are ranked next .
3. Second best performing EOS in the prediction of Cp are
BWRS and BWR-LS, while CAP, AGA8 and SRK are
ranked next .
AGA8, while CPA, SRK and PR are ranked next .

8 DISCUSSION
To give some perspective of the order of magnitude impact
the errors in the three thermodynamic properties evaluated in
the present work, a buried NPS 48 pipeline of length = 70 km,

W.T. = 25.4 mm, transporting an average mixture of NG of
MW= 17.617 blended with 20%(mole) hydrogen such that the
blended mixture MW = 14.496, is considered. The internal
wall roughness parameter = 15 m. The upstream pressure
(discharge pressure from an upstream compressor station) is
7.8 MPa, and temperature is 30oC. The overall heat transfer
coefficient between the pipe external wall and the ground is
1.81 W/m2.oC, and the ground temperature is 1 oC. The
downstream pressure (i.e. at inlet to the next compressor
station) is 5.8 MPa. Based on these conditions and applying
GERG2008 EOS for density and enthalpy calculations, the
predicted flow rate is 619.0981 kg/s, and the downstream gas
temperature is calculated to be 18.087 oC.
If another EOS is used such that its density prediction is of
the order of ±1%, the flow throughput would be in error by
±0.5%. Here the effect of density is accounted for in the
momentum equation including effects of Reynold number on
the friction factor according to Colebrook-White correlation
[63]. If on the other hand, changes in enthalpy incur an error
of the order of ±1%, it will result in an error in the downstream
temperature of the order of -/+0.053 oC (i.e. in the opposite
direction), albeit very small. Error in the isentropic enthalpy
rise across a compressor (head) and hence power is directly
proportional to the error in predicting Cp. That is, a ±1% error
in Cp results in the same ±1% error in the enthalpy rise across
a compressor and power consumed. Additionally, a ±1% error
in Cp results in ±0.69% error in the temperature rise across the
compressor in the same direction. Finally, an error of the order
of ±1% in the speed of sound will result in a proportional error
and in the same direction in the gas impedance and hence in
pressure perturbations due to unsteady and transient flows.
The EOS ranking in this paper is based purely on the
relative error from measured data; the best equation of state to
use in an actual application may have other considerations
such as what is readily available in the software being used,
does the increased accuracy in the end result justify additional
computational effort to calculate EOS, how much uncertainty
is there in the other variables that go into the analysis, how
flexible are downstream processes to account for variances in
the EOS, etc.
9 CONCLUSIONS
The following overall conclusions can be drawn from the
present work:
Figure 13: Performance Ranking of the 12 EOS in Terms 1. Overall, GERG2008 performed best in predicting the
of Density, Speed of Sound and Heat Capacity three principal thermodynamic properties: density, speed
Prediction Errors for Binaries and NG Mixtures with of sound and isobaric heat capacity for pure H2 and
Hydrogen in the Range of 10-25 MPa. binaries and NG mixtures with H2 in all ranges of
pressure and temperature.
2. The second-best performing EOS is AGA8 for pure H2.

3. The second-best performing EOS is BWRS and AGA8 for Measurement Committee Report No. 8, AGA Catalog No. XQ
Binaries and NG + H2 mixtures in the range of pressure 1285, Arlington, VA.
of 0.1-3 MPa, and 3-25 MPa, respectively. This is based [5] Factors for Natural Gas and Other Related Hydrocarbon Gases",
on the overall error criterion defined by Eq. (4), which is American Gas Association, Transmission Measurement
Committee Report No. 8.
the sum of the average of the absolute errors in these three [6] Starling, K. E. and Savidge, J. L. (1992). "Compressibility

thermodynamic properties. Starling, K. E. (1971). "Thermo Data Refined for LPG",
4. Other EOS were found to be ranked next to the second Hydrocarbon Processing, 3, 101.
best depending on which thermodynamic properties are of [7] ISO-12213-2, 1997, Natural gas—Calculation of compression
interest. Table 7 summarizes these options which are factor—Part 2: Calculation using molar-composition analysis,
given the term “acceptable”. More specifics can be ISO, Ref. No. ISO- 12213-2:1997(E).
obtained from respective ranking charts corresponding to [8] Benedict, Manson; Webb, George B.; Rubin, Louis C. (1940),
each row in Table 7. These are Figs. 8, 10, 11, 12 and 13, "An Empirical Equation for Thermodynamic Properties of Light
respectively. Hydrocarbons and Their Mixtures: I. Methane, Ethane, Propane,
and n-Butane", Journal of Chemical Physics, 8 (4): 334–345,
Bibcode:1940JChPh...8.334B, doi:10.1063/1.1750658, ISSN
0021-9606.
Table 7: Overall Performance Ranking of the Second [9] Starling, Kenneth E. (1973), Fluid Properties for Light
Best and “Acceptable” EOS. Note that GERG2008 is Petroleum Systems, Gulf Publishing Company, p. 270, ISBN
found to be the Best Overall in All Pure and NG 978-0872012936
Mixtures. [10] Brule, M.R., Lin, C.T., Lee, L.L. and Starling, K.E., 1982.
Multiparameter corresponding‐states correlation of coal‐fluid
Pressure Range Temperature
Total #
Isobaric Heat
thermodynamic properties. AIChE Journal, 28(4), pp.616-625.
of Data Density Speed of Sound Overall (E )
Mixtures:
(MPa) Range (oC)
Points
Capacity (Cp) [11] Peng, D. Y. and Robinson, D. B. (1976). "A New Two-Constant
Min Max Min Max #
Second
Best EOS
Acceptable*
Second
Best EOS
Acceptable*
Second
Best EOS
Acceptable*
Second
Best EOS
Acceptable* Equation of State", Ind. Eng. Chem. Fundam., 15(1), 59-64.
PR-MHV2, PR-MHV2,
CPA, BWR- [12] Huron, M.J. and Vidal, J., 1979. New mixing rules in simple
Pure H2 0.1 50 -75 75 629 RKS-BM SRK RKS-BM LS and RKS-BM SRK
PR-BM, SRK PR-BM
BWRS equations of state for representing vapour-liquid equilibria of
Binary and
AGA8, CPA,
PR-BM, PR-
BWR-LS, CPA, PR,
strongly non-ideal mixtures. Fluid Phase Equilibria, 3(4),
NG Mixtures 0.1 50 -100 200 1788 SRK AGA8 MHV2, PR- BWRS AGA8
with H2
RKS-BM, PR
WS, PR
CPA SRK
pp.255-271.
Sub-sets of the Above Binary and NG Mixtures Sub-sets of the Above Binary and NG Mixtures
[13] Boston, J.F. and Mathias, P.M., 1980, March. Phase equilibria
Binary and
in a third-generation process simulator. In Proceedings of the
PR-BM, PR-MHV2, BWRS, BWR- CAP, AGA8, CPA, SRK,
NG Mixtures
with H2
10 25 -20 80 523 SRK AGA8, PR
AGA8 PR-WS, PR LS SRK
AGA8
PR 2nd international conference on phase equilibria and fluid
Binary and BWR-LS,
properties in the chemical process industries (pp. 823-849).
BWR-LS, AGA8, PR- PR-WS, PR- BWRS, CPA,
NG Mixtures 3 10 -20 80 517 SRK
BWRS BM MHV2, PR
BWRS CPA, AGA8,
SRK, PR
AGA8
PR Great Neck, NY: Deutsche Gesellschaft für Chemisches
with H2
Binary and RKS-

Apparatewesen.
NG Mixtures 0 3 -20 80 536 BWRS, SRK
BM,CPA,B
WR-LS,
AGA8
BWR-LS,
BWRS
BWRS, BWR-
LS
CPA, SRK
BWRS, BWR- CPA, AGA8,
LS SRK [14] Wong, D.S.H. and Sandler, S.I., 1992. A theoretically correct
with H2 AGA8
* Average of the absolute errors for these EOS are slightly higher than the second best EOS that could be
mixing rule for cubic equations of state. AIChE Journal, 38(5),
acceptable for certain applications
pp.671-680.
[15] Soave, G., 1972. Equilibrium constants from a modified
Redlich-Kwong equation of state. Chemical engineering
ACKNOWLEDGEMENTS science, 27(6), pp.1197-1203.
This paper is part of a research program sponsored by TC [16] Redlich, Otto.; Kwong, J. N. S. (1949-02-01). "On the
Energy, and permission to publish is gratefully acknowledged. Thermodynamics of Solutions. V. An Equation of State.
Fugacities of Gaseous Solutions". Chemical Reviews. 44 (1):
233–244. doi:10.1021/cr60137a013. ISSN 0009-2665.
[17] Michels, A., De Graaff, W. and Wolkers, G.J., 1964.
REFERENCES Thermodynamic properties of hydrogen and deuterium at
[1] Kunz, O. and Wagner, W., 2012, "The GERG-2008 Wide- temperatures between− 175 C and 150 C and at pressures up to
Range Equation of State for Natural Gases and Other Mixtures: 2500 atmospheres. Applied Scientific Research, Section A,
An Expansion of GERG-2004", J. Chem. Eng. Data, 57 (11), pp 12(1), pp.9-32.
3032-3091. [18] Michels, A., De Graaff, W. and Wolkers, G.J., 1959.
[2] Kontogeorgis, Georgios M.; Michelsen, Michael L.; Folas, Thermodynamic properties of hydrogen and deuterium at
Georgios K.; Derawi, Samer; von Solms, Nicolas; Stenby, temperatures between− 175° C and 150° C and at densities up to
Erling H. (2006). "Ten Years with the CPA (Cubic-Plus- 840 amagat. Physica, 25(7-12), pp.1097-1124.
Association) Equation of State. Part 1. Pure Compounds and [19] Johnston, H.L. and White, D., 1950. Pressure-volume
Self-Associating Systems". Industrial and Engineering temperature relationships of gaseous normal hydrogen from its
Chemistry Research. 45 (14): 4855–4868. boiling point to room temperature and from 0–200 atmospheres.
doi:10.1021/ie051305v. Trans. Am. Soc. Mech. Engrs, 72, pp.785-7.
[3] Kontogeorgis, Georgios M.; Voutsas, Epaminondas C.; [20] FORREZ, G. and DROOGMANS, G., 1960.
Yakoumis, Iakovos V.; Tassios, Dimitrios P. (1996). "An MEASUREMENTS ON THE VELOCITY OF SOUND IN
Equation of State for Associating Fluids". Industrial & GASEOUS NORMAL HYDROGEN UP TO 75 ATM. A. VAN
Engineering Chemistry Research. 35 (11): 4310–4318. ITTERBEEK, W. VAN DAEL. Bulletin de l'Institut
[4] AGA 8, 1992, Compressibility and super compressibility for international du froid, 40, p.91.
natural gas and other hydrocarbon gases, Transmission

[21] Michels, A., De Graaff, W., Wassenaar, T., Levelt, J.M.H. and [41] Heath, H. R. Proc. Phys. Soc., Sect. B, 1953, 66, 362 The
Louwerse, P., 1959. Compressibility isotherms of hydrogen and Viscosity of Gas Mixtures.
deuterium at temperatures between-175C and+ 150C (at [42] Gururaja, G. J.; Tirunarayanan, M. A.; Ramachandran, A. J.
densities up to 960 amagat). Physica, 25(1-6), pp.25-42. Chem. Eng. Data, 1967, 12, 562-567 Dynamic Viscosity of Gas
[22] Machado, J. R. S.; Streett, W. B.; Deiters, U. J. Chem. Eng. Data, Mixtures
1988, 33, 148-152 PVT Measurements of Hydrogen/Methane [43] Wormald, C. J. ELDATA: The International Electronic Journal

Mixtures at High Pressures of Physico-Chemical Data, 1995, 1, 95-112 Selected
[23] Mason, D. M.; Eakin, B. E. J. Chem. Eng. Data, 1961, 6, 499- experimental high temperature-high pressure excess enthalpy
504 Compressibility factor of fuel gases at 60 F and 1 atm data for 16 binary systems containing steam
[24] Mihara, S.; Sagara, H.; Arai, Y.; Saito, S. J. Chem. Eng. Jpn., [44] Bartlett, E. P. J. Am. Chem. Soc., 1927, 49, 1955-1957 The
1977, 10, 395-9 The compressibility factors of hydrogen - Compressibility Isotherms of Hydrogen, Nitrogen and Mixtures
methane, hydrogen - ethane and hydrogen - propane gaseous of These Gases at 0 deg and Pressures to 1000 Atmospheres. A
mixtures. Correction
[25] Magee, J. W.; Pollin, A. G.; Martin, R. J.; Kobayashi, R. Fluid [45] Bartlett, E. P.; Hetherington, H. C.; Kvalnes, H. M.; Tremearne,
Phase Equilib., 1985, 22, 155-173 Burnett-isochoric P-V-T T. H. J. Am. Chem. Soc., 1930, 52, 1363-1373
measurements of a nominal 20 mol% hydrogen-8 mol% [46] Michels, A.; Wassenaar, T. Appl. Sci. Res., 1949, 1, 258-62
methane mixture at elevated temperatures and pressures Isotherms of argon + nitrogen + hydrogen mixture between 0
[26] Solbrig, C. W.; Ellington, R. T. Chem. Eng. Prog. Symp. Ser., and 150 c up to 340 atmospheres
1963, 59, 127-136 The p-v-t behavior of selected hydrogen- [47] Jaeschke, M.; Hinze, H. M.; Achtermann, H. J.; Magnus, G.
methane and hydrogen-ethane mixtures. Fluid Phase Equilib., 1991, 62, 115-139 PVT data from Burnett
[27] Jett, M.D. Ph.D. Thesis, Rice University, Houston (1990) and refractive index measurements for the nitrogen-hydrogen
Isochoric PVT Studies of Binary Fluid Mixtures: The Hydrogen- system from 270 to 253 k and pressures to 30 mpa
Methane and Methane-Methanol Systems [48] Verschoyle, T. T. H. Proc. R. Soc. London, Ser. A, 1926, 111,
[28] Jett, M.; Fleyfel, F.; Kobayashi, R. J. Chem. Thermodyn., 1994, 552 Isotherms of hydrogen of nitrogen, and of hydrogen +
26, 561-569 The nature of (p,Vm,T) surfaces of mixtures: nitrogen mixtures, at 0 and 20 C, up to a pressure of 200
(quantum + normal) and ( normal + hydrogen-bonded), as atmospheres
summarized by their isochoric inflection loci [49] Deming, W. E.; Shupe, L. E. J. Am. Chem. Soc., 1931, 53, 860
[29] Wormald, C. J.; Lewis, K. L.; Mosedale, S. E. J. Chem. The beattie-bridgeman equation of state and bartlett's p-v-t data
Thermodyn., 1977, 9, 27-42 on a 3:1 hydrogen + nitrogen mixture
[30] Golubev, I. F.; Gnezdilov, N. E. Teploenergetika, 1967, 14, 93- [50] Bennett, C. O.; Dodge, B. F. Ind. Eng. Chem., 1952, 44, 180
4 Viscosity of methane-hydrogen mixtures at 273-523.degree.K. Compressibilities of mixtures of hydrogen and nitrogen above
under pressures up to 490.3 .times. 105 newtons/m.2 1000 atmospheres
[31] Chuang, S-.y; Chappelear, P. S.; Kobayashi, R. J. Chem. Eng. [51] Gray, P.; Wright, P. G. Proc. 2nd Symp. Thermophys. Prop.,
Data, 1976, 21, 403-411 Viscosity of Methane, Hydrogen, and 1962, 35, 395-403, Thermal Conductivities and Wassiljewa
Four Mixtures of Methane and Hydrogen from -100 C to 0 C at Coefficients for Gaseous Mixtures of Ammonia and Its
High Pressures Elements
[32] Trautz, M.; Sorg, K. G. Ann. Physik, 1931, 10, 81-96 Viscosity, [52] Saxena, S. C.; Gupta, G. P. J. Chem. Eng. Data, 1970, 15, 98-
heat conductivity and diffusion in gas mixtures. XVI. The 107 Experimental Data and Procedures for Predicting Thermal
viscosity of ~ u hydrogen, methane, ethane, propane and their Conductivity of Multicomponent Mixtures of Nonpolar Gases
binary mixtures [53] Saxena, S. C.; Tondon, P. K. J. Chem. Eng. Data, 1971, 16, 212-
[33] Adzumi, H. Bull. Chem. Soc. Jpn., 1937, 12, 199 Studies on the 220 Experimental Data and Procedures for Predicting Thermal
Flow of Gaseous Mixtures Through Capillaries I. The Viscosity Conductivity of Binary Mixtures of Nonpolar Gases
of Binary Gaseous Mixtures [54] Gray, P.; Wright, P. G. Proc. R. Soc. London, Ser. A, 1961, 263,
[34] Zhang, J.; Zhang, X.; Han, B.; He, J.; Liu, Z.; Yang, G. J. 161-188 The thermal conductivity of mixtures of nitrogen,
Supercrit. Fluids, 2002, 22, 15-19 Study on intermolecular ammonia and hydrogen
interactions in supercritical fluids by partial molar volume and [55] Boyd, J. H. Phys. Rev., 1930, 35, 1284 The Viscosity of
isothermal compressibility Compressed Gases
[35] Cipollina, A.; Anselmo, R.; Scialdone, O.; Filardo, G.; Galia, A. [56] Van Itterbeek, A.; van Paemel; van Lierde Physica
J. Chem. Eng. Data, 2007, 52, 2291-2297 (Amsterdam), 1947, 13, 88 :MEASUREMENTS ON THE
[36] Souissi, M. A. B.; Kleinrahm, R.; Yang, X.; Richter, M. J. VISCOSITY OF GAS MIXTURES
Chem. Eng. Data, 2017, 62, 2973-2981 [57] Van Itterbeek, A.; Van Paemel, O.; VanLierde, J. Physica
[37] Ababio, B. D.; McElroy, P. J. J. Chem. Thermodyn., 1993, 25, (Amsterdam), 1947, 13, 231-239 Measurements on the Thermal
1495-1501 (Pressure, amount-of-substance density, Diffusion in Gas Mixtures at Low Temperatures
temperature) of carbon dioxide + hydrogen {(1-x)CO2+xH2} [58] Pal, A. K.; Barua, A. K. J. Chem. Phys., 1967, 47, 216-218
using a direct method Viscosity of Hydrogen Nitrogen and Hydrogen Ammonia Gas
[38] Bezanehtak, K.; Combes, G. B.; Dehghani, F.; Foster, N. R.; Mixtures
Tomasko, D. L. J. Chem. Eng. Data, 2002, 47, 161-168 [59] Kestin, J.; Yata, J. J. Chem. Phys., 1968, 49, 4780-4791
[39] Ibbs, T. L.; Hirst, A. A. Proc. R. Soc. (London), 1929, 123, 134- Viscosity and Diffusion Coefficient of Six Binary Mixtures
142 The Thermal Conductivity of Gas Mixtures [60] Lozano-Martin, D., Martín, M.C., Chamorro, C.R., Tuma, D.
[40] Maltsev, V. A.; Nerushev, O. A.; Novopashin, S. A.; and Segovia, J.J., 2020. Speed of sound for three binary (CH4+
Radchenko, V. V.; Licht, W.; Miller, E. J.; Parekh, V. S. J. H2) mixtures from p=(0.5 up to 20) MPa at T=(273.16 to 375)
Chem. Eng. Data, 2004, 49, 684-687 Viscosity of H2-CO2 K. International Journal of Hydrogen Energy, 45(7), pp.4765-
Mixtures at (500, 800, and 1100) K 4783.

[61] Lozano-Martín, D., Vega-Maza, D., Moreau, A., Martín, M.C., Thermodynamic and Transport Properties - REFPROP,
Tuma, D. and Segovia, J.J., 2021. Speed of sound data derived Version 9.0, Gaithersburg: National Institute of Standards
perfect-gas heat capacities, and acoustic virial coefficients of a and Technology, Standard Reference Data Program,
calibration standard natural gas mixture and a low-calorific H2- 2010.
enriched mixture. The Journal of Chemical Thermodynamics,
[63] Colebrook, C. F., and White, C. M., 1937, "Experiments
158, p.106434.

with Fluid Friction in Roughened Pipes," Proc. Lond. R.
[62] Lemmon, E.W., Huber, M.L. and McLinden, M.O., NIST
Soc., A 161, PP. 367-378.
Standard Reference Database 23: Reference Fluid
Appendix:
Table 1: Data of Measured Properties of Pure, Binary and Mixtures of NG with Hydrogen.
# of Data # of Data
# of Data Total #
Pressure Range Temperature Points for Points for
o Components Points for of Data
(MPa) Range ( C) Soeed of Isobaric Heat
Mixtures: Density
Sound Capacity (Cp)
Points
Min Max Min Max # # # #
Pure H2 0.1 50 -75 75 100% H2 386 159 84 629
Binary and H2:N2,CO2,C1,C2,C3,

NG Mixtures 0.1 50 -100 200 iC4,nC4,iC5,nC5, 1506 282 none 1788
with H2 neoC5,C6,He, H2O,O2
Sub-sets of the Above Binary and NG Mixtures

NG Mixtures 0.1 3 -20 80 iC4,nC4,iC5,nC5, 479 57 none 536

NG Mixtures 3 10 -20 80 iC4,nC4,iC5,nC5, 434 83 none 517

NG Mixtures 10 25 -20 80 iC4,nC4,iC5,nC5, 402 121 none 523
Note: For the "none' cells above, GERG2008 EOS was taken as reference to compare other EOS to.

Performanceof

Uploaded by

Copyright:

Available Formats

Performanceof

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Performanceof

Uploaded by

Copyright:

Available Formats

Proceedings of the 2022 14th International Pipeline Conference

Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

Kamal K. Botros Larry Jensen

1 Copyright © 2022 by ASME

Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

2 Copyright © 2022 by ASME

Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

Figure 2: Data of Measured Speed of Sound for Pure

Figure 1: Data of Measured Densities for Pure Hydrogen

Figure 3: Data of Measured Isobaric Heat Capacity for

3 Copyright © 2022 by ASME

4 Copyright © 2022 by ASME

Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

to low pressure distribution systems. Results are

3 RESULTS FOR PURE HYDROGEN

5 Copyright © 2022 by ASME

# Average of Absolute Error (%) Sum of Absolute

Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

1 GERG2008 0.033 0.213 0.151 0.397

2 CPA 2.190 1.490 0.855 4.535

3 AGA8 2.640 2.780 1.040 6.460

4 BWRS 4.630 4.890 0.948 10.468

5 BWR-LS 4.520 4.620 0.931 10.071

6 PR 1.770 1.510 1.550 4.830

7 PR-MHV2 2.610 1.310 0.571 4.491

8 PR-BM 2.610 1.310 0.571 4.491

9 PR-WS Poor Performance 2.620 -

10 SRK 1.220 1.320 0.971 3.511

11 RKS-BM 0.530 0.625 1.020 2.175

12 RKS-WS Poor Performance

Figure 7: Example of Density, Speed of Sound and Heat

6 Copyright © 2022 by ASME

Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

7 Copyright © 2022 by ASME

Table 3: Summary Performance Table of the 12 EOS in

Binary & NG Mixtures with H2

# Average of Absolute Error (%) Sum of Absolute

1 GERG2008 0.084 0.039 0.062 0.185

2 CPA 0.871 0.933 0.439 2.243

3 AGA8 0.836 0.453 0.546 1.835

4 BWRS 1.700 2.530 0.321 4.551

5 BWR-LS 1.590 3.210 0.376 5.176

6 PR 0.886 0.824 0.687 2.397

7 PR-MHV2 2.090 0.664 1.900 4.654

8 PR-BM 1.530 0.468 0.940 2.938

8 Copyright © 2022 by ASME

Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

AGA8 are ranked next .

9 Copyright © 2022 by ASME

Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

Figure 11: Performance Ranking of the 12 EOS in Terms

10 Copyright © 2022 by ASME

Downloaded from http://asmedigitalcollection.asme.org/IPC/proceedings-pdf/IPC2022/86564/V001T08A002/6965742/v001t08a002-ipc2022-86297.pdf by China University of Petroleum user on 23 November 2023

# Average of Absolute Error (%) Sum of Absolute

1 GERG2008 0.128 0.036 0.082 0.246

2 CPA 0.574 0.841 0.457 1.872

3 AGA8 0.572 0.332 0.555 1.459

4 BWRS 0.547 0.944 0.251 1.742

5 BWR-LS 0.485 1.390 0.365 2.240

6 PR 0.711 0.625 0.611 1.947