The project goal is the development of reliable methods for predicting macroscopic properties that are important for describing the flow storage of one or more fluid phases in reservoirs at an intrinsic scale approaching that of a local volume average or representative element volume using nuclear magnetic resonance (NMR).
This project was selected in response to DOE's Oil Exploration and Production solicitation DE-RA26-98BC15200.
Texas A&M University
College Station, TX
The detailed knowledge of rock and fluid properties is essential to the success of petroleum reservoir management and characterization. However, the study of heterogeneous media has been limited by the lack of methods to spatially resolve properties within porous media. Conventional methods utilize inflow and outflow measurements, and often do not adequately resolve heterogeneities.
NMR spectroscopy and magnetic resonance imaging (MRI) can give noninvasive measurements within media. Suitable interpretation of the data provides unprecedented opportunities for resolving fluid states to determine macroscopic properties important for describing the flow of one or more fluid phases in reservoirs.
In Stage I, the project developed advanced core analysis tools to determine macroscopic properties-porosity and absolute and relative permeability-within heterogeneous core samples. In Stage II, the project used these methods, together with additional NMR spectroscopic measurements, to obtain data for development of predictive methods. In Stage III, the project developed improved relations for predicting permeabilities and testing a novel method for predicting relative permeability from NMR well-log observable properties.
The new method for the estimation of surface relaxivity enables the researcher to include the effect of a distribution of pore sizes in the model. As a result, it gives improved estimates of surface relaxivity and the corresponding pore-size distributions. Pore size distribution is valuable for determining the flow capacity of more than one fluid phase within a reservoir.
Methods for determining the surface relaxivity from measured data were developed and tested with data obtained from samples provided by ExxonMobil. The new method avoids the use of a certain mathematical short-time approximation in the data analysis, which has been shown to be unsuitable.
Numerical work was completed to simulate two-phase displacement experiments using all three spatial dimensions when estimating two-phase flow functions relative to permeability and capillary pressure properties. The modification of the computer core was completed, including tests for the changes observed in specific examples.
The computer code SENDRA, previously used to simulate two-phase displacement experiments, is limited to one or two spatial-dimension problems. Numerical work has extended the simulator to three-dimensional applications to simulate three-dimensional problems; thus the source code of SENDRA has been modified. The final step was the integration of the modified code with SENDRA and testing on three-dimensional cases.
The project is complete.
$202,302 (20% of total)