Forecasting Oil Formation Volume Factor for API Gravity Ranges Using Artificial Neural Network

Abstract

The Oil Formation Volume Factor (FVF) parameter is a very important fluid property in reservoir engineering computations. Ideally, this property should be obtained from actual measurements. Quite often, this measurement is either not available, or very costly to obtain. In such cases, empirically derived correlations are used in the prediction of this property. This work centers on building an artificial neural network (ANN) model to predict oil formation volume factor for the different API gravity ranges. The new models were developed using combination of 448 published data from the Middle East, Malaysia, Africa, North Sea, Mediterranean basin, Gulf of Persian fields and 1389 data set collected from the Niger Delta Region of Nigeria. The data have been divided into the following four different API gravity classes: heavy oils for API≤21, medium oils for 21 35. The data set was randomly divided into three parts of which, 60% was used for training, 20% for validation, and 20% for testing for each particular API grade. Both quantitative and qualitative assessments were employed to evaluate the accuracy of the models to the existing empirical correlations. The ANN models outperformed the existing empirical correlations by the statistical parameters used with the best rank and better performance plots. Key words : Oil formation volume factor; Artificial neural network; Back propagation; Statistical analysis; API gravity ranges