Simulation of Compositional Processes: The Use of Pseudocomponents in Equation-of-State Calculations
- A.G. Schlijper (Koninklijke/Shell E and P Laboratorium)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- September 1986
- Document Type
- Journal Paper
- 441 - 452
- 1986. Society of Petroleum Engineers
- 5.5 Reservoir Simulation, 5.2 Reservoir Fluid Dynamics, 5.2.2 Fluid Modeling, Equations of State, 5.8.8 Gas-condensate reservoirs, 5.2.1 Phase Behavior and PVT Measurements
- 0 in the last 30 days
- 243 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 12.00|
|SPE Non-Member Price:||USD 35.00|
Simulation of Compositional Processes: The Use of Processes: The Use of Pseudocomponents in Pseudocomponents in Equation-of-State Calculations
Summary. A new method is presented for performing equation-of-state (EOS) calculations with pseudocomponents that are obtained by lumping pure components of a multicomponent reservoir fluid description. The method recognizes that a pseudocomponent represents a mixture of components and thus has the thermodynamic behavior of a mixture, rather than that of a single component. The standard expression for the chemical potential is replaced by an adapted pseudopotential. The method also yields convenient procedures for obtaining phase densities and viscosities. procedures for obtaining phase densities and viscosities. The accuracy of the method is demonstrated by comparing phase splits, lumped mole fractions in liquid and vapor phases, and phase densities and phase viscosities obtained from a lumped fluid description with the same phase viscosities obtained from a lumped fluid description with the same data as those obtained from a frilly compositional description of hydrocarbon systems. Generally, deviations are not more than a few percent. The method enables the use of a reduced compositional fluid percent. The method enables the use of a reduced compositional fluid representation in numerical simulation studies without any essential loss in modeling accuracy.
Analytic EOS'S, particularly cubic ones, are widely used in the oil industry to predict the phase behavior of reservoir fluids. An accurate description of the thermodynamics of a reservoir oil by means of such an EOS, however, generally requires the use of many components. In compositional numerical reservoir simulators that use an EOS, the computing time and storage requirements for a simulation increase sharply with the number of components involved and soon become prohibitively large. The conflict between the goals of speed of computation and accuracy of modeling can present a serious problem, particularly in the simulation of miscible drives where particularly in the simulation of miscible drives where proper modeling of compositional effects is essential. proper modeling of compositional effects is essential. To reduce a detailed description of a fluid system from a relatively large to a small number of components, the concept of the pseudocomponent has been introduced: a group of actual components is considered as one pseudocomponent. Although a pseudocomponent is thus a pseudocomponent. Although a pseudocomponent is thus a mixture of components rather than one (pure) component, reported methods for the lumping of reservoir fluid components invariably treat pseudocomponents the same as pure components in the EOS calculations 1-5 : a pure components in the EOS calculations 1-5 : a pseudocomponent is treated as a hypothetical pure component pseudocomponent is treated as a hypothetical pure component characterized by parameters that are some kind of average of the characterizing parameters of the constituents. We shall refer to this method as the averaged-component method or averaged-component approximation.
In this paper, we propose a new method for the use of pseudocomponents in EOS calculations. In this method, pseudocomponents in EOS calculations. In this method, which we shall refer to as the pseudopotential method, a pseudocomponent is not characterized or treated in the same way as a pure component, but its mixture-like character is accounted for. This shows up in that, for a pseudocomponent, an adapted, pseudoized version of the pseudocomponent, an adapted, pseudoized version of the chemical potential is used, which shall be called a pseudopotential. pseudopotential. In addition to providing pseudopotentials to replace the standard chemical potentials in the vapor/liquid equilibrium (VLE) equations, the pseudopotential method supplies accurate estimates of phase compositions in terms of the many original pure components. Thus it is possible to use well-established compositional correlations to possible to use well-established compositional correlations to calculate phase properties, and resorting to often-arbitrary mixing-rule approaches is not necessary.
The pseudopotential method is based on the assumption that a reservoir fluid can be described as a mixture of weakly polydisperse fluids.
In VLE calculations for an nc-component fluid system at fixed pressure, p, and temperature, T, the thermodynamic quantity of interest is the molar Gibbs free Energy, G, of the system, which is the sum of contributions from the liquid phase, subscript L, and the vapor phase, subscript V: phase, subscript V: (1)
Here xi (yi) denotes the mole fraction of Component i in the liquid (vapor) phase, and fL (fv) is the molar liquid (vapor) fraction.
|File Size||901 KB||Number of Pages||12|