36
10–20 components representing individual compounds (e.g., CH 4 or CO 2 ) or group
of poorly identified compounds (e.g. C 10 –C 14 , all compounds between normal decane and normal tetradecane). Chemical properties of these petroleum fractions
(including critical properties, acentric factors, and molecular weight) need to be
collected from literature or estimated (Riazi 2005). Predictions of equations of state
are usually validated with laboratory data obtained with reservoir fluid samples, and
equation of state “tuning” is also a frequently used procedure to improve prediction
ability when enough data is available (Zick 2013). This tuning consists of a careful
modification of pseudo-component properties or binary interaction parameters performed to minimize the discrepancy with available laboratory data. Predictions of
selected properties are discussed in turn below.
3.3.1 Bubble Point
The bubble point can be predicted from EOSs using published procedures (e.g.
(Michelsen and Mollerup 2007)). This calculation is relatively straightforward at
low-pressure conditions. However, a careful procedure at higher pressures is necessary to avoid convergence of the algorithm to a false solution.
3.3.2 Gas Saturation
Gas saturation of a petroleum mixture can be calculated with an equation of state.
The amount of “gas” (e.g., CH 4 ) is gradually increased until the equation of state
predicts two phases rather than one.
At high-volume ratios of gas to petroleum liquid, the gas phase will deplete the
petroleum liquid from its lighter components, affecting the equilibrium content of
light C 1 –C 5 components within the petroleum liquid. Equation of state calculations for
such systems highlight the importance of taking this effect into consideration when
designing laboratory measurement procedure. Figure 3.6 shows solubility data determined by the two methods described in Sect. 3.2.2 along with modeled data.
3.3.3 Density and Swelling
Density of both gas and petroleum liquid phases are readily predicted from cubic
equations of states, though they are known to lead to poor predictions of the densities
of liquid petroleum phases (Ahmed 2010). However, the so-called “volume translation” procedures have been developed to improve prediction abilities of equations of
states (Lin and Duan 2005; Péneloux et al. 1982). Very good prediction abilities for
binary mixtures can be achieved when using volume translation (e.g., Fig. 3.7).
T. B. P. Oldenburg et al.
10–20 components representing individual compounds (e.g., CH 4 or CO 2 ) or group
of poorly identified compounds (e.g. C 10 –C 14 , all compounds between normal decane and normal tetradecane). Chemical properties of these petroleum fractions
(including critical properties, acentric factors, and molecular weight) need to be
collected from literature or estimated (Riazi 2005). Predictions of equations of state
are usually validated with laboratory data obtained with reservoir fluid samples, and
equation of state “tuning” is also a frequently used procedure to improve prediction
ability when enough data is available (Zick 2013). This tuning consists of a careful
modification of pseudo-component properties or binary interaction parameters performed to minimize the discrepancy with available laboratory data. Predictions of
selected properties are discussed in turn below.
3.3.1 Bubble Point
The bubble point can be predicted from EOSs using published procedures (e.g.
(Michelsen and Mollerup 2007)). This calculation is relatively straightforward at
low-pressure conditions. However, a careful procedure at higher pressures is necessary to avoid convergence of the algorithm to a false solution.
3.3.2 Gas Saturation
Gas saturation of a petroleum mixture can be calculated with an equation of state.
The amount of “gas” (e.g., CH 4 ) is gradually increased until the equation of state
predicts two phases rather than one.
At high-volume ratios of gas to petroleum liquid, the gas phase will deplete the
petroleum liquid from its lighter components, affecting the equilibrium content of
light C 1 –C 5 components within the petroleum liquid. Equation of state calculations for
such systems highlight the importance of taking this effect into consideration when
designing laboratory measurement procedure. Figure 3.6 shows solubility data determined by the two methods described in Sect. 3.2.2 along with modeled data.
3.3.3 Density and Swelling
Density of both gas and petroleum liquid phases are readily predicted from cubic
equations of states, though they are known to lead to poor predictions of the densities
of liquid petroleum phases (Ahmed 2010). However, the so-called “volume translation” procedures have been developed to improve prediction abilities of equations of
states (Lin and Duan 2005; Péneloux et al. 1982). Very good prediction abilities for
binary mixtures can be achieved when using volume translation (e.g., Fig. 3.7).
T. B. P. Oldenburg et al.