In this work, we present a field study which indicates that if PVT properties are properly derived, we may be able to use a beta model to history match and predict performance for complex fluid systems. The specific example considered pertains to a field containing a gas condensate fluid column underlaid by a volatile oil leg. Residue gas and nitrogen are continuously injected into four wells completed in the gas condensate column, whereas, 29 wells produce from the oil zone. For this system, we were able to build a two-component beta model that accurately reproduced the same thirty years of reservoir performance as an eleven component compositional simulator originally formulated for history matching and performance prediction.
The intended major contributions of this work are as follows. First, we present a procedure which uses a calibrated equation of state combined with material balance techniques to generate appropriate PVT properties for the beta model. The PVT properties properly account for the variation in saturation pressure from one point to another in the reservoir. Second, we illustrate that a beta model can often be used for history matching and performance for cases where it is commonly believed that a multicomponent compositional model is needed.