An Intelligent Approach for Permeability Prediction in a Carbonate Reservoir: (Case Study of an Iranian Reservoir)

In this study available routine core and wireline logs data from five wells are used to develop a model for prediction of permeability. In the first step, Artificial Neural Network (ANN) is used to predict permeability but, the results were not sufficiently accurate. Therefore, the concept of hydraulicflow units is taken into consideration to overcome the problem ofheterogeneity. Cluster analyses are used to provide the optimal number of flow units that exist in the formation. In the next step for the purpose of flow unit prediction, fuzzy logic approach is used. Then ANN is used for prediction of permeability in each specific HFU. Obtained results show thatthe predicted values are very close to the actual data with high accuracy. The results of permeability prediction based on this technique were highly satisfactory when predictive capability of models were examined in thecored wells. Therefore, in the light of reliable results, flow units are extrapolated to an uncored but logged well and synthetic permeability log is generated for this well.