| dc.description.abstract | It is believed that the observed rapid rise in global temperatures is caused by high atmospheric concentration of CO2, due to emissions from fossil fuel combustion. While global efforts are currently in place to mitigate the effect, it is expected that hydrocarbons will remain the main source of energy supply for the planet in the foreseeable future. Harmonizing these seemingly conflicting objectives has given rise to the concept of Carbon Capture Utilization and Storage (CCUS).
A prominent form of CCUS involves the capture and injection of anthropogenic CO2 for Enhanced Oil Recovery (EOR). During CO2 EOR, substantial amount of injected CO2 is retained and permanently stored in the subsurface. However, due to inherent geological and thermodynamic complexities in subsurface environments, most CCUS projects are plagued with poor sweep efficiencies. For successful CCUS implementation, advanced reservoir management strategies which appropriately capture relevant physics are therefore required. In this regard, effective techniques in three fundamental areas of reservoir management including forward modeling, inverse modeling and field development optimization methods are presented herein. In each area, we demonstrate the validity and utility of our methodologies for CCUS applications with field examples.
First, a comprehensive streamline-based simulation of CO2 in saline aquifers is proposed. Here, the unique strength of streamlines at resolving sub-grid resolution which enables a high-resolution representation of CO2 transport during injection is exploited. Relevant physics such as compressibility and formation dry-out effects which were
ignored in previously proposed streamline models are accounted for. The methodology is illustrated with a series of synthetic models and applied to the Johansen field in North Sea. All streamline-based models are benchmarked with commercial compositional simulation response with good agreement.
Second, a Multiresolution Grid Connectivity-based Transform (M-GCT) for effective subsurface model calibration is proposed. M-GCT allows the representation and update of grid property fields with improved spatial resolutions. This enables improved characterization of the subsurface, especially for CCUS systems in which CO2 transport is highly sensitive to contrasts in hydraulic conductivity. The approach is illustrated with a synthetic and a field scale problem. To demonstrate its utility, the proposed method is applied to a field actively supporting a post-combustion CCUS project.
Finally, a streamline-based rate optimization of intelligent wells used in CCUS projects is proposed. Based on a previously developed method, a combination of the incremental oil recovery, CO2 storage efficiency and CO2 utilization factor are optimized through an optimal rate schedules of the installed ICVs. The approach is particularly efficient since required objective function gradients and hessians are computed analytically from streamline-derived sensitivities obtained from a single simulation run. This significantly reduces the computational expense required to obtain solutions at level of optimality comparable to existing methods. The approach is illustrated with a synthetic case and applied to the Norne field to demonstrate the robustness of the approach. | en |
