INTEGRATED RESERVOIR PROPERTY ESTIMATION USING ARTIFICIAL NEURAL NETWORK OVER ‘FREEMAN’ FIELD, OFFSHORE NIGER DELTA

Abstract

Reservoir properties have a significant impact on petroleum field operations and reservoir management. Porosity and Permeability are two fundamental characteristics of reservoir systems that are typically distributed in a spatially non-uniform and non-linear manner. Hence, they are difficult to estimate in a reservoir when or where there is limited information. The research aims at developing a comprehensive and accurate estimate of reservoir properties in ‘Freeman’ field, offshore Niger Delta from seismic, digital well log and core data using the Artificial Neural Network (ANN) approach. Artificial neural network was used as a nonlinear regression method to develop transformation between the selected well logs and core analysis data as a means to estimating reservoir properties in ‘Freeman’ field. The Neural network estimation was carried out using PETREL™ and MATLAB™ software packages. Five well log responses - Gamma Ray (GR), Deep Resistivity (RD), Formation Density (DEN), Neutron Porosity (PHIN) and Interval transit Time (ΔT) were inputs in training the ANN for estimation, while core porosity and permeability were used as target data. Correlation between estimated values of porosity and permeability, core porosity and permeability were given as 0.9997 and 0.8784 respectively, and this shows satisfactory results from the ANN. Also, petrophysical model providing a geologically realistic spatial porosity and permeability distribution by showing the variation in values of these properties within the reservoir was made. This research was able to show that ANN can be used to estimate reservoir properties in and away from wells from seismic, well and core data in order to get a better understanding of the subsurface reservoir heterogeneity within the ‘Freeman’ field of the Niger Delta.