Application Of Artificial Neural Networks To Predict Wettability And Relative Permeability Of Sandstone Rocks

Abstract

An Artificial Neural Network (ANN) model based on the back-propagation technique is trained with a number of variables from experimentally established relative permeability curves. The reservoir core input data covers an extensive range of porosities and permeabilities from different sandstone lithologies having diverse wettabilities. The trained model is then tested with only a couple of input variables such as the initial connate water saturation, S,»c and the residual oil saturation. So, . The developed model outputs, or the predictions define the relative permeability end-points and the intersection point to quantify the wettability and the shape of the relative permeability curves. A number of correlations based on empirical models and network models exist to predict the relative permeability curves and the wettability of oil bearing sandstone formations from the initial oil and water. Calculations from the ANN model were then compared with values calculated from other models currently in wide spread use.