Effects of Recent Relative Permeability Data on CO2 Flood Modeling

Abstract

Abstract Results from laboratory tertiary CO2 flooding studies conducted at representative reservoir conditions are becoming available. Predicted CO2 flood performance can be significantly changed by using this data in reservoir models. The laboratory data includes water and oil relative permeabilities when the water saturation is decreasing, residual oil saturations to a miscible flood, residual CO2 saturations, and CO2 relative permeabilities. Predicted oil recovery, CO2 production, and breakthrough times are all influenced. Unfortunately, much of this data cannot be used in most presently available commercial reservoir simulators. Additionally, there is uncertainty about the proper form for the relative permeability of the miscible (non-aqueous) phase. For example, should the miscible phase relative permeability be based on the gas, oil, or solvent relative permeability or some combination? Texaco has developed a mixing parameter based reservoir simulator that not only uses the new relative permeability data but also incorporates different forms for the miscible phase relative permeability. This paper describes how the recently available laboratory data and the form of the miscible relative permeability formulation affect predicted CO2 flood performance. Results presented show the importance of using the laboratory data and show what changes in existing simulators may be advisable.