Assessment of Clustering Methods for Predicting Permeability in a Heterogeneous Carbonate Reservoir

Abstract

Permeability, the ability of rocks to flow hydrocarbons is directly determinedfrom core. Due to high cost associated with coring, many techniques suggested to predict permeability from easy to obtain and frequent properties of reservoir such as porosity. This study launched to put clustering methods (MRGC, SOM, DC and AHC) into practice in permeability prediction of a heterogeneous carbonate reservoir in SW Iran. Conventional approaches like empirical models and regression analysis were also examined. Performance of all examined methods compared in order to choose the best approach for predicting permeability in uncored wells of the studied field. For all clustering methods, selecting optimal number of clusters is the most important task. The optimal value for the number of clusters selected by iteration. The optimal number of clusters for MRGC, SOM, DC and AHC is 7, 9, 9 and 8 respectively. Empirical equations and regression analysis could not predict permeability well and R2 of estimation of both approach were around 0.6. Generally performance of clustering techniques was acceptable regarding to Fahliyan formation. These techniques predict permeability between 1 and 1000 mD very well and just overestimate permeability below 1 mD. SOM performed the best between examined techniques (R2=0.7911).