Determination of Reservoir Electrofacies Using Clustering Methods (MRGC, AHC, SOM,DYNCLUST) Throughout Arab Part in Salman Oil Field 2S-05 Well

Primary objective of this project is introducing the best clustering method to determine the Electrofacies in without core wells. Electrofacies concept in fact is a deterministic or analytical method for clustering petrophysical well-log data. They can indicate changes in geological features or reservoir. Electrofacies is defined based on clustered data, and placement of logs and similar data in the same group which is differentiated from other groups. In this study, different clustering methods were studied and compared with each other. Among these methods, the best method was introduced as MRGC method. For this purpose, the Geolog software was employed to use four clustering methods including Multi Resolution Graph based Clustering method (MRGC), Ascendant hierarchical method (AHC), Self-organizing neural networks method (SOM) and dynamic clustering method (DYNCLUST) to determine the reservoir electrofacies in Salman 2S-05 well. Thus, 9 electrofacies were determined in each clustering. Electrofacies produced by the best clustering method (MRGC), using well logs including GR, DT, RHOB and NPHI. They were sorted based on reservoir quality from good to poor. This study was performed on the Arab member. Arab member is the main reservoir in Salman field, and the highest oil production (70%) from this formation has been reported.