Loading...
CO₂ flooding impact on compositional changes and their effect on elastic properties in oil reservoirs
Dashti, Abdullah
Dashti, Abdullah
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2024
Date Submitted
Keywords
Collections
Files
Loading...
Dashti_mines_0052N_12942.pdf
Adobe PDF, 5.91 MB
Dashti_mines_0052N_316/inverse 34 API2 MISCIBLE_SGMOD_injectivitynoDTcompressible.dat
Unknown, 33.68 KB
Dashti_mines_0052N_316/inverse 34 API2 IMMISCIBLE_SGMOD_injectivitynoDTcompressible.dat
Unknown, 33.64 KB
Dashti_mines_0052N_316/inverse 24 API2 lowerPV_SGMOD_injectivitynoDTcompressible.dat
Unknown, 28.37 KB
Dashti_mines_0052N_316/inverse 24 API2 IMMISCIBLE_SGMOD_injectivitynoDTcompressible.dat
Unknown, 28.4 KB
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
This research aims to enhance our understanding of the CO2 interactions with the reservoir oil and brine and, consequently, improve the methods for monitoring and tracking CO2 distribution within the reservoir via P-wave velocity responses in a seismic survey. The research was motivated by the growing importance of CO2 utilization for enhanced oil recovery (EOR) and subsequent sequestration. As for the outcome, this study quantifies the effect of CO2 mixing with reservoir fluids on the elastic reservoir properties to further evaluate the hypothesis from a previous study (Oduwole 2022) that showed that not accounting for compositional changes in oil during CO2 injection would lead to an underestimation of the magnitude of the bulk modulus of the actual response.
This study uses a compositional reservoir simulator software, GEM (from Computer Modelling Group, CMG) to simulate CO2 injection and its subsequent effects on the fluid and petrophysical properties, specifically, the reservoir’s bulk modulus and acoustic velocity. This research employed two different fluid systems to compare a heavier oil composition with a lighter composition—both in miscible and immiscible conditions to determine the effect of fluid segregation during flooding. Additionally, the study compares two different approaches that implement Gassmann fluid substitution calculations to estimate bulk modulus and acoustic velocity based on the compositional reservoir simulation results: a black oil approach of Gassmann fluid substitution calculations and a compositional approach.
As for numerical modeling and solution, several issues need careful attention. The first one is the fact that CO2 dissolves in brine substantially, and the second item is the grid-orientation effect on the position and distribution of the fluid-displacing front. For instance, in this thesis, a nine-point finite difference approach was implemented instead of a five-point finite difference in the solution of the flow equations. The latter distorts the injection fluid front due to the higher mobility of the displacing phase in comparison with reservoir fluids. In addition, accounting for CO2 solubility in brine delays the displacing front and reduces the concentration of CO2 dissolved in hydrocarbon phases.
The findings of this study show that in a three-phase system, the velocity change is predominately influenced by the supercritical CO2 phase. However, the supercritical CO2 phase’s density and bulk modulus are significantly influenced by the components from the oil phase. Heavier oils result in the CO2 absorbing more heavy components, whereas lighter oils lead to the supercritical CO2 absorbing more lighter components. Therefore, the supercritical CO2 phase density varies from the injection site to the production as it mobilizes different components from the oil phase.
The changes in saturated bulk modulus, acoustic P-wave velocity, and acoustic impedance are initially significantly different when the Gassmann black oil approach is compared to the compositional approach. These differences are more pronounced for miscible conditions than for immiscible conditions. This is a result of the supercritical CO2 mixing within the oil phase at the displacement front, which the black oil Gassmann approach is incapable of capturing. This leads to an underestimation of the location of the front that is behind the actual front in years for the miscible model and months for the immiscible. Additionally, due to the increasing purity of the injected CO2 as time elapses, the models start aligning with each other despite the changes in the oil density and bulk modulus.
Associated Publications
Rights
Copyright of the original work is retained by the author.