Search for a command to run...
Abstract In the SAGD process, steam injection may result in pore pressure changes around the steam chamber in the reservoir. The magnitude of the pore pressure change depends on the steam injection pressure and the relative position of the steam chamber within the reservoir. Injected steam also increases the temperature within and adjacent to the steam chamber and thus introduces thermal expansion effects. The complex interaction of pore pressure and temperature throughout the reservoir result in varying degrees of reservoir geomechanical interactions that must be considered when contemplating reservoir geomechanical simulations of the SAGD process. The primary geomechanical influence on SAGD recovery is associated with the volume change of the sand matrix in response to effective stress changes induced by steam injection pressures and temperatures. Reservoir parameters and processes, such as compressibility, porosity (pore volume), absolute permeability, relative permeability, saturations, capillary pressure, enthalpy transmissibility, gas evolution, and thermal expansion effects, are all affected by bulk volume changes. In this paper, variations of these parameters due to geomechanical effect are analyzed and discussed based on related test results, calculation, and simulation. Their impacts on reservoir geomechanical simulations of the SAGD process are discussed. Introduction Reasonable prediction of SAGD performance by numerical simulation is an integral component in the design and management of a SAGD project. Conventional reservoir numerical simulation emphasizes multiphase flow in the porous media but generally does not take the interactions between fluid and solid into account. It applies elastic rock compressibility to characterize the coupling mechanism of multiphase flow and rock skeleton. The assumption of this treatment is that the boundary loads and temperature are constant, Δp = Δ σ'(1, 2). All analytical flow equations in petroleum engineering are based on this assumption. It is clear that the recovery process of conventional oil from sandstones and most carbonate rocks can roughly satisfy this assumption. For the SAGD process, however, volumetric deformations within the reservoir due to pore pressure and temperature changes result in variations of both in situ stress and strain. These stress and strain variations are functions of the in situ boundary conditions. Due to different directional deformation in response to in situ heating, the total stresses in vertical and horizontal directions may also vary. Moreover, the SAGD process is mainly applied in friable or uncemented (unconsolidated) oil sands that geomechanically behave differently to their cemented counterparts. Under these conditions, the assumption used in conventional numerical simulation is no longer effective because the total stress changes within the reservoir.