Bubble Point Pressure Prediction Model for Niger Delta Crude using Artificial Neural Network Approach

Abstract Knowledge of bubble point pressure is very important in reserve estimation and other petroleum engineering calculations such as modeling of fluid flow through porous media and multiphase flow in pipes. Usually, this property is obtained from laboratory PVT analysis. However, when such analysis is not available, empirically derived PVT correlations are used. This work focuses on the use of an Artificial Neural Network (ANN) to address the inaccuracy of empirical correlations used for predicting bubble point pressure. The ANN is a mathematical model inspired by biological neural networks. In this modeling approach 1248 data sets collected from the Niger Delta Region of Nigeria were used. The data set was randomly divided into three parts, of which 60% was used for training, 20% for validation, and 20% for testing. The accuracy of the new Artificial Neural Network was compared with existing empirical correlations. The ANN model outperformed the existing empirical correlations by the statistical parameters used with a best rank of 17.3132 and better performance plot.