PVT Properties and Viscosity Correlations for Gulf of Mexico Oils
- Birol Dindoruk (Shell Intl. E&P) | Peter G. Christman (Shell Intl. E&P)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Evaluation & Engineering
- Publication Date
- December 2004
- Document Type
- Journal Paper
- 427 - 437
- 2004. Society of Petroleum Engineers
- 4.1.9 Tanks and storage systems, 5.5 Reservoir Simulation, 5.2 Reservoir Fluid Dynamics, 4.1.5 Processing Equipment, 4.1.2 Separation and Treating, 5.4.2 Gas Injection Methods, 5.8.7 Carbonate Reservoir, 4.6 Natural Gas, 5.2.1 Phase Behavior and PVT Measurements, 5.5.1 Simulator Development
- 2 in the last 30 days
- 1,630 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 12.00|
|SPE Non-Member Price:||USD 35.00|
New empirical pressure/volume/temperature (PVT) correlations for Gulf of Mexico (GOM) oils have been developed as a function of commonly available field data. Correlations have been developed for the following:
- Bubblepoint pressure.
- Solution gas/oil ratio (GOR) at bubblepoint pressure.
- Oil formation volume factor (FVF) at bubblepoint pressure.
- Undersaturated isothermal oil compressibility.
- Dead-oil viscosity.
- Saturated-oil viscosity.
- Undersaturated-oil viscosity.
For the development of correlations, we have covered a wide range of data. More than 100 PVT reports from the GOM have been used in the development of correlations. Two of the published correlations (those of Standing1 and Petrosky and Farshad2) were tested with our GOM data set. Proposed correlations of this study predicted the PVT properties of GOM oils better than the correlations published in the literature, even when the coefficients of the published correlations are tuned. Using the correlations of this study, we have written a simple program that can generate PVT data sets for use in reservoir simulation. Laboratory-measured differential-liberation data for two different oil samples were also compared against our empirically generated differential-liberation data.
In reservoir studies, from material-balance calculations to simulation, fluid properties are always necessary to estimate the in-place volumes, surface volumes, and transport pmeters that interact with the flow. The variations of PVT properties during the depletion phase are also needed to evaluate the reservoir performance and to design surface and subsurface facilities.
Ideally, PVT properties are experimentally measured in the laboratory. When such direct measurements are not available, PVT correlations from the literature are often used. Fundamentally, there are two different types of correlations in the literature. The first group of correlations is developed with randomly selected data sets; we will refer to such correlations as generic correlations. The second group of correlations is developed using a known geographical area or a certain class/type of oil. Correlations using randomly selected data sets may not be suitable for certain types of oils or for some geographical areas. Even though the authors of the generic correlations attempt to cover a wide range of data, such correlations still work better for certain types of oils. Specialized correlations represent the properties of a certain type of oil or geographical area (for which they are developed) better than the generic correlations.
Crudes are complicated hydrocarbon mixtures, and their detailed composition may not always be available. Even when there is detailed compositional information available, it is common practice to group the components into pseudocomponents (lumping). Black-oil representation of oils is a special case of lumping in which the reservoir fluid is represented by a septor-gas component and a stock-tank-oil component. Black-oil PVT data often can be correlated with practically measurable quantities such as oil gravity (API), gas gravity, solution GOR, and reservoir temperature. Note that representation of oils with only a few pmeters may not always be enough to characterize their behavior within an acceptable tolerance. For example, a naphthenic oil may have the same API, solution GOR, and gas gravity as its pffinic counterpart. Although they may have totally different viscosities, most correlations would predict the same viscosity for those two oils.
We have tested the correlations published in the literature against the available GOM data. Our work indicates that the accuracy and the ranges of validity of those published correlations are not suitable for our GOM resource base (especially for the deepwater applications). Therefore, there is a need for a suite of new PVT correlations for GOM oils.
All fluid samples were obtained from reservoirs in the GOM. We have used approximately 100 PVT laboratory reports. Reservoir temperature (Tk), single-stage flash data for solution GOR (Rs), gas gravity (?g), and oil gravity (API), were used for the development of the correlations. Statistical distributions of the input data are shown in Tables 1 and 2. As shown in Table 1, the bubblepoint pressure ranged between 926 and 12,230 psia. Corresponding solution GORs ranged from 133 to 3,050 scf/STB. According to McCain,3 3,050 scf/STB is close to the limit (3,300 scf/STB) at which the reservoir fluid is retrograde gas at reservoir conditions. Similar to pbp and solution GOR, oil FVF at the bubblepoint pressure (Bobp) varied between 1.08 and 2.9 RB/STB. We also have used septor-oil FVFs to develop a relationship between differential-liberation values (BobpDL) and the septor-adjusted values (Bobpsep) for septor temperature correction (see Appendix A). The range of the septor-oil FVFs (1.08 to 2.74 RB/STB) is similar to the range of differential-liberation oil FVFs. However, septor-oil FVFs are either equal to or lower than the differenial-liberation oil FVFs. As expected from the ranges of pbp and Bobp, bubblepoint oil-compressibility range was also wide (5 to 32 psi-1).
Compared to our work, Petrosky and Farshad4 used data covering smaller ranges of pbp (1,574 to 6,523 psia), Rs (217 to 1,406 scf/STB), Bobp (1.12 to 1.62 RB/STB), and co (4 to 25 psi-1). However, viscosity ranges considered for the viscosity correlations were similar to the ones proposed here. Some of the PVT reports used in this study did not include viscosity measurements at desired conditions, and some reports exhibited dead-oil viscosities (outliers) that were too high. After discarding the outliers and the incomplete data sets, more than 90 PVT reports (see Table 2 for details) were used for the development of the viscosity correlations.
|File Size||1 MB||Number of Pages||11|