This paper describes a numerical study evaluating a gas re-injection process as a method of recovering additional oil from thin (less than 5 m) heavy oil reservoirs that have experienced primary production. The reservoir data was provided by Fletcher Challenge Energy Canada Ltd. The target reservoirs were the Lloydminster Dulwich and Buzzard reservoirs. The process concept was to re-energise the reservoir by injecting CH4, CH4/C3H8, or CO2, using a horizontal well, or an existing vertical well plus a wormhole network. The oil recovery process based on foamy oil expansion, oil dilution and viscosity reduction must be employed. Available oil properties and geological data, including relative permeability curves, were evaluated and were used as inputs to the CMG STARS numerical simulator. A satisfactory match of primary production history was used as a template to predict the proposed injection and post-primary production. The oil production for various re-injection options was compared to determine the feasibility of the process.
Modest incremental CH4anges in performance were noted for both reservoirs. The numerical simulations predicted the 90/10 C3H8/CH4 mix was the best performer for the Dulwich Reservoir, and the 20/80 C3H8/CH4 mixture was the best solvent of those evaluated for the Buzzard Reservoir.
The objectives of this work were to provide a comparative study of gas recharge options for thin reservoirs of various types, and to provide field support for Fletcher Challenge Energy Canada ltd. (FCEC), by simulating the process based on data from 2 reservoirs. The reservoirs were the Dulwich reservoir, a 10 m thick, 1200 mPa.s oil, 4 Darcy heavy oil reservoir with 26 years of production history (Table 1), and the Buzzard reservoir, a 6 m thick, 1 Darcy reservoir with 10,000 mPa.s oil, with 16 years of production history (Table 2). The numerical simulation work was done using the CMG STARS numerical simulator. PVT calculations were done using CMG's Winprop PVT simulator.
The simulation work consisted of simulations to history-match field production data, and comparative simulations of post-primary process options for the Dulwich and Buzzard reservoirs. The simulations also examined parameter sensitivities and a number of implementation options. The simulations were initially done with a 2D radial grid mesh in order to enable the completion of a large number of quick scoping simulations. The scoping simulation results were then used to select simulation cases to be done using a 3D Cartesian grid. The same procedure was followed for both the Dulwich and Buzzard reservoir systems.
The gas recharge process is a complex post-primary oil recovery process, depending for its success on the interplay of several mechanisms. Primary production is by foamy oil drive, accompanied in some cases by sand production and wormhole propagation. During gas injection, the gas is distributed in the reservoir by following wormhole paths, and by viscous fingering. Diffusion and hydrodynamic dispersion also play a role in the movement of gas into the reservoir oil. During post-primary production, oil is driven to the producing well by the gas drive, by renewed foamy oil drive, and in some cases by existing natural drives (water drive).