A five-year laboratory and simulation investigation on heavy-oil solution-gas drive processes is reported. It is shown how the interpretation of these processes has changed over the years. Initially, it was attempted to take into account the "bubbly" character of the oil by changing the oil thermodynamical properties. Non-conventional PVT properties accounted for supersaturation and trapped gas bubbles. Later, in agreement with work in other research groups, the low produced gas-oil ratio and high recovery factors were modeled by introducing low gas-oil relative permeability. This allowed for history matching of laboratory and field data, but doubts remain about the predictive capacity of reservoir simulation models. This doubt is motivated by the observed flow geometry of the (dispersed flow), which is not physically taken into account in the reservoir simulation equations. It is concluded that more work is necessary on the determination of the mechanisms that govern the flow, and on the development ofm odels that more correctly represent the physics of dispersed gas-liquid flow in porous media. Further experiments in cores and micromodels, together with microscopic modeling of flow in pore spaces, should be part this work.

You can access this article if you purchase or spend a download.