Reservoir Permeability Prediction Using Artificial Neural Network; A Case Study of “XZ” Field, Offshore Niger Delta

No Thumbnail Available
Ozebo, V.C
Ezimadu, C.C
Journal Title
Journal ISSN
Volume Title
Akamai University, U.S.A
Reservoir Permeability is one of the most important characteristics of hydrocarbon bearing formations. A good knowledge of a formation’s permeability helps geophysicist to efficiently manage the production process. Formation permeability is often measured in the laboratory from cores or evaluated from well test data. Core analysis and well test data, however, can only be gotten from a few wells in a field due to economic factors, while majority of wells are logged. In this study, an artificial neural network has been designed with PETRELTM, which is able to predict permeability of a formation using the data gotten from geophysical well logs with good accuracy. A case study from XZ field offshore Niger Delta is presented. Five well log responses (Gamma Ray Log (GR), Deep Resistivity (RD), Formation Density (DEN), Neutron Porosity (PHIN) and Density Porosity (PHID)) were initially used as inputs in the ANN to predict permeability. Core permeability from one of the wells (OS1) was used as target data in the ANN to test the prediction. The accuracy of the ANN approach is tested by regression plots of predicted values of permeability with core-permeability which is the standard. Excellent matching of core data and predicted values reflects the accuracy of the technique. Permeability estimations/predictions presented in this paper have a correlation coefficient of 0.8 where 1.0 is a perfect match. This work showed that prediction result is improved by adding core porosity in the training, carefully selecting input data and increasing the number of iterations reasonably.
artificial neural networks, reservoir; permeability, petrel, Niger delta
Ozebo, V.C. and C.C. Ezimadu. (2019). “Reservoir Permeability Prediction Using Artificial Neural Network; A Case Study of “XZ” Field, Offshore Niger Delta”. Pacific Journal of Science and Technology. 20(2):309-318