Permeability Estimation Using an Integration of Multi-Resolution Graph-based Clustering and Rock Typing Methods in an Iranian Carbonate Reservoir

Rock typing has been utilized in numerous studies where it has been proven to be a powerful tool for determining rock properties and estimating unknown parameters such as permeability. It can be performed based on routine core analysis (RCAL) or special core analysis (SCAL) data, and the accuracy of results could be different. Because of the high cost and time-consuming process of special core analysis, SCAL data are not available in all wells of a reservoir. Hence, in this study, a practical workflow is carried out using RCAL data. For this purpose, the data of four wells in a reservoir have been used. After utilizing three HFU (Hydraulic Flow Units), Winland r35 and lithology methods, the results showed that the best and the most accurate rock typing method is Winland r35 method. In the next step, several approaches were used to estimate permeability, and it was observed that the combination of the multi-resolution graph-based clustering (MRGC) method in GEOLOG software and Winland r35 method in this carbonate reservoir is the best estimation approach. The correlation coefficient (R2), between measured and estimated permeability was approximately 0.96. Eventually, when the only available data are the RCAL data, the presented algorithm yields a high degree of accuracy.