Abstract
The Port Neches CO2 miscible flood project began CO2 injection in September, 1993 into a waterflooded sandstone reservoir along the Texas Gulf Coast. Sponsored by the U.S. Department of Energy (DOE) in their Class I Oil Program, this project will determine the recovery efficiency of CO2 flooding a sandstone reservoir which has been extensively waterflooded down to a residual oil saturation of 30%. The design of this project utilized the various tools available for predicting the recovery performance of such projects, with DOE's CO2 Predictive Model CO2PM1 and a compositional model being used. In addition a streamtube model2,3 has been developed to predict the recoveries associated with the waterflood and CO2 recovery processes. The validity of this streamtube model, the CO2PM program, and previous compositional reservoir simulation work, has been evaluated by the use of a compositional five-spot model where an equation-of-state for the current reservoir oil is incorporated. This work points out the streamtube model's ability as an effective screening device for CO2 flood prediction. Furthermore, the importance of properly characterizing the permeability within each layer of the reservoir is demonstrated by the improved recoveries seen in fining-upward sequence reservoirs.