Page 88 - Fundamentals of Enhanced Oil and Gas Recovery
P. 88

76                                                                      Ramin Moghadasi et al.


                3.2.2.1.6 Rostami et al. [42]
                Recently, Rostami et al. [42] have proposed two new and accurate models for predic-
                tion of CO 2 solubility. In their research, they examined different correlations over a
                wide range of experimental data. In the second step, they developed two models using
                neural networks and gene expression programming (GEP). In a comparative study,
                they concluded that GEP is much more accurate than any other correlation both for
                live and dead crude oils. They have proposed this method as a feasible approach for
                CO 2 flooding simulations. Their proposed correlations are as follows:
                •  Dead oils:
                      Based on the latest investigations, they considered that the key variables influ-
                   encing CO 2 solubility in dead oil are oil molecular weight (MW), specific gravity
                   of oil (γ), reservoir temperature (T), and saturation pressure (P s ). The CO 2 solubil-
                   ity in dead oils can be calculated by the following equation:

                                               ð
                                           P s T 5:6444 1 0:008756MWÞ
                              R s 5                                       γ            (3.9)
                                          2
                                                                 2
                                   8:9318P 1 0:010819MWP s T 1 T 1 41:105 T
                                          s
                •  Live oil:
                      For the case of live oils, they considered that CO 2 solubility is primarily depen-
                   dent on oil molecular weight (MW), specific gravity of oil (γ), reservoir temperature
                   (T), saturation pressure (P s ), and bubble point pressure (P b ). Eq. (3.10) has been
                   developed by Rostami et al. [42] for predicting CO 2 solubility in live oils as follows:
                                          7:3695P b 2 7:3713P s 1 0:48618
                              R s 5                                                   (3.10)
                                   0:021262MW 1 4:6233P b 2 5:0337P s 2 γT 2 A
                      In which the parameter A is a conditional function, which is defined by the
                   following relationship:

                                           0              if γ # 0:849
                                     A 5                                              (3.11)
                                           0:042756MW     if γ . 0:849
                      For both cases, the units for MW, T, P s , P b , and R s are g/mole, C, MPa, MPa,

                   and mole fraction, respectively.


                3.2.2.2 Swelling Effects
                This effect is the most obvious effect that gas injection could have on oil recovery dur-
                ing an immiscible process. When a reservoir oil is not saturated with gas or reservoir
                pressure increases due to gas injection so that more gas can be dissolved, the volume of
                gas dissolved in oil will increase until the oil is saturated at that pressure. As this phe-
                nomenon occurs, oil formation volume factor (FVF) will increase. This phenomenon is
                called oil swelling, which can significantly increase oil recovery. Swelling effects are less
                significant for those reservoirs with a gas cap; however, for the reservoir oils that do not
   83   84   85   86   87   88   89   90   91   92   93