Predicting correlations properties of crude oil systems using type-2 fuzzy logic systems

This paper presented a new prediction model of pressure–volume–temperature (PVT) properties of crude oil systems using type-2 fuzzy logic systems. PVT properties are very important in the reservoir engineering computations, and its accurate determination is important in the primary and subsequent development of an oil field. Earlier developed models are confronted with several limitations especially in uncertain situations coupled with their characteristics instability during predictions. In this work, a type-2 fuzzy logic based model is presented to improve PVT predictions. In the formulation used, the value of a membership function corresponding to a particular PVT properties value is no longer a crisp value; rather, it is associated with a range of values that can be characterized by a function that reflects the level of uncertainty. In this way, the model will be able to adequately model PVT properties. Comparative studies have been carried out and empirical results show that Type-2 FLS approach outperforms others in general and particularly in the area of stability, consistency and the ability to adequately handle uncertainties. Another unique advantage of the newly proposed model is its ability to generate, in addition to the normal target forecast, prediction intervals without extra computational cost.

► We presented a new prediction model of pressure–volume–temperature (PVT) properties of crude oil systems using type-2 fuzzy logic systems. ► Earlier models are confronted with several limitations especially in uncertain situations coupled with their characteristics instability during predictions. ► In the proposed model, the value of a membership function corresponding to a particular PVT properties value is not a crisp value; rather, it is associated with a range of values that can be characterized by a function that reflects the level of uncertainty. ► In this way, the model will be able to adequately model PVT properties. The model also generates prediction intervals without extra computational cost. ► Empirical results show that this approach outperforms others in general and particularly in the area of stability, consistency and the ability to adequately handle uncertainties.