oa Reservoir Porosity and Permeability Prediction from Petrographic Data Using Artificial Neural Network - A Case Study from Saudi Arabia

Understanding reservoir heterogeneity is essential for the assessing and the prediction of the reservoir<br>properties and quality. This study investigates the prediction of the reservoir petrophysical properties<br>of the Ordovician Upper Dibsiyah Member of the Wajid Sandstone in south west Saudi Arabia. The<br>Artificial Neural Networks (ANNS) technique is used here to study the pattern recognition and<br>correlation among the petrographic thin section data such as grain size, sorting, matrix % and<br>cementation % and perophysical properties of the reservoir such as porosity, permeability and lithofacies.<br>For this purpose, artificial intelligence techniques were designed and developed and these are the<br>multilayer perception (MLP) and the general regression neural network (GRNN). The good agreement<br>between core data and precdicted values by neural netwoks demonstrate a successful implementation<br>and validation of the network’s ability to map a complex non-linear relationship between petrographic<br>data, permeability and porosity. The GRNN technique provides better prediction of the reservoir<br>properties than that obtained from the use of the MLP technique.