HR: 1330h
AN: GC32A-0201    [PDF]
TI: Rapid Prediction of CO2 Movement in Aquifers, Coal Beds, and Oil and Gas Reservoirs
AU: Orr, F M
EM: fmorr@stanford.edu
AF: Petroleum Engineering Dept Stanford University, 074 Green Earth Sciences Bldg 367 Panama Street, Stanford, CA 94305 United States
AU: * Jessen, K
EM: krisj@pangea.stanford.edu
AF: Petroleum Engineering Dept Stanford University, 074 Green Earth Sciences Bldg 367 Panama Street, Stanford, CA 94305 United States
AU: Kovscek, A
EM: kovscek@pangea.stanford.edu
AF: Petroleum Engineering Dept Stanford University, 074 Green Earth Sciences Bldg 367 Panama Street, Stanford, CA 94305 United States
AB: Predictions of the mix of future primary energy sources often include significant use of fossil fuels, and scenarios envisioning a switch to renewable and/or nuclear primary energy sources rely on fossil fuels for the extended period required to install large-scale systems. Effective means of sequestering CO2 will be required to reduce emissions of CO2 in these scenarios. The earth's crust presents three major classes of geologic formation that appear suitable for long-term storage: deep formations containing salt water, unmineable coalbeds, and depleted oil and gas reservoirs. With injection into oil and gas reservoirs and coalbeds, it may be possible to recover net energy in concert with CO2 storage. If CO2 injection into geologic formations is undertaken on a large scale, high-resolution, but low computational cost, numerical methods will be needed. Such simulations may be used to predict where CO2 is likely to flow, interpret the volume and spatial distribution of the subsurface contacted by injectant, and optimize injection operations. These elements will certainly be necessary if geological sequestration is proven feasible and public acceptance is to be gained. In this paper, we present research on developing ultra-fast computational methods and tools applicable to the suite of geologic formations suitable for CO2 storage. The underpinnings of these methods are streamline-based computations. The flow field in 3D is decoupled into a series of 1D flow problems linked by common injection and boundary conditions. Periodically, streamline trajectories are updated as the pressure field in the volume under consideration evolves. The advantages of this approach are a reduction in the dimensionality of the numerical problem, the possibility to employ analytical solutions along each streamline, and a significant reduction in the effects of numerical dispersion. In contrast, conventional finite-difference based numerical techniques suffer from excessive numerical dispersion and long computation times. Finally, we demonstrate by calculation examples the different mechanisms controlling the displacement behavior of CO2 sequestration schemes, the interaction between flow and phase equilibrium and how proper design of injection gas composition and well completion are required to co-optimize oil production and CO2 storage.
DE: 1694 Instruments and techniques
DE: 1699 General or miscellaneous
SC: Global Climate Change [GC]
MN: 2003 Fall Meeting