Porosity Prediction of Unayzah Reservoir from Well Log Data Using Backpropagation Neural Network

By

Mahbub Hussain¹, Aamir Siddiqui¹, Abdulazeez Abdulraheem¹, Gabor Korvin¹

(1) King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia

 Unayzah reservoir in Saudi Arabia is a Permian, siliciclastic reservoir and it is the source of light sweet crude oil (Arabian Super light) and gas.

This paper reports the findings of application of Backpropagation Neural Network for the prediction of porosity values of Unayzah reservoir in Haradh Field using genetic approach. The usage of genetic approach involves the classification of well log data into various lithofacies groups and then porosity prediction were carried out on a facies-by-facies basis.

Results from the present study showed that Backpropagation neural network provides reasonable prediction accuracy and predicted porosity values are highly correlated with neutron porosity values.