Solution gas/oil ratio (GOR) and oil formation volume factor (FVF) are normally obtained from differential or flash liberation tests. However, neither the differential liberation process nor the flash liberation process can represent the fluid flow in petroleum reservoirs. Therefore, data obtained from any of the two test procedures must be adjusted to approximate the fluid behavior in the reservoir. At low pressures, the conventional method of adjustment yields negative values of solution GOR and values of oil FVF of less than 1. This, of course, is not physically correct.
This paper presents a new method for adjusting the differential liberation data to separator conditions. The new method overcomes the limitations of the conventional adjustment method and makes the low-pressure extension of the curves of solution GOR and oil FVF more accurate. The method is based on the fact that data obtained from both the differential and flash liberation tests should yield the same value of oil relative density at reservoir conditions. The new method is tested using 425 PVT files, yielding results that are consistent with the physical behavior of solution GOR and oil FVF.
In the differential liberation process, gas is removed as it is released from the oil. In the flash liberation process, however, the liberated gas is not removed and is allowed to reach equilibrium with the oil.
Generally, petroleum engineers consider that the gas liberation process in the reservoir can be represented by the differential liberation process.1,2 The fluid produced from the reservoir to the surface is considered to undergo a flash process.
The differential solution GOR is not the same as the flash solution GOR, as shown in Fig. 1. Similarly, the differential and flash oil FVFs are not the same, as depicted in Fig. 2. Thus, regardless of the testing procedures - flash or differential - some correction needs to be made to the obtained data to approximate the fluid behavior in the oil-production process.
The actual gas liberation process in the reservoir is neither flash nor differential. In certain localities, the process is flash, but in others, the process is differential. In some other localities the process does not match either of them. A combination test proposed by Dodson et al.3 is probably the closest to the reservoir process. At each step of the differential liberation test, an oil sample is taken and flashed to obtain Rs, ?o, Bo, and ?g. Here it can be seen that all properties, including the ?api, are different at different pressures. Although this combination test is an improvement over the differential and flash liberation tests, it does not match the actual reservoir behavior. The appendix to Ref. 4 explains the differential and flash processes and the combination test. From the combination test that produces different values for ?g and ?o at different pressures, it is justified to adjust all the properties obtained by the differential liberation test to flash separator conditions, including ?g and ?o.
The fluid properties obtained by combining data from the differential and flash liberation tests may be called the "combination fluid properties." These data are used to determine the values of solution GOR and oil FVF at pressures below bubblepoint. To calculate the combination fluid properties from standard data analysis, several assumptions were stipulated, but these assumptions limit the range of application.
This paper describes a new method to adjust the differential liberation data to separator conditions. This method overcomes the disadvantages and limitations of the existing method and comes up with a correction procedure that results in a consistent physical trend.
The existing method for adjusting the differential liberation data to separator conditions was based on several assumptions, the most important of which (as stipulated by Amyx et al.5) are:
The standard cubic feet of gas remaining in solution at reservoir conditions that will be liberated upon producing that liquid to the separator by a flash liberation process is the difference between the original gas in solution and the differentially liberated gas corrected for the reservoir shrinkage of the fluid.
The relationship between the oil FVFs of flash and differentially separated samples remains constant over the entire pressure range of interest.
In equation form, the corrected differential solution GOR at pressures below bubblepoint pressure, according to the first assumption mentioned above, is as follows:
where RS=solution GOR adjusted to separator conditions; Rsbf=bubblepoint solution GOR obtained from the separator test; Rsbd=bubblepoint solution GOR obtained by the differential liberation test; Rsd=differential solution GOR; Bobf =bubblepoint oil FVF flashed through the separator to stock-tank conditions; and Bobd=bubblepoint oil FVF differentially liberated to stock-tank conditions.
Implicitly, the adjusted differential solution GOR at pressures above the bubblepoint pressure is constant and is equal to the solution GOR at the bubblepoint obtained from the separator test.
According to the second assumption, the adjusted differential oil FVF at pressures below the bubblepoint pressure is evaluated from a combination of differential liberation data and separator test data; that is,
where Bod=oil FVF obtained by differential liberation tests, and Bo=oil FVF adjusted to separator conditions.
Implicitly, the adjusted differential oil FVF at bubblepoint pressure is equal to the oil FVF at bubblepoint pressure obtained from the separator test; that is,