Abstract
To delineate a powerful reservoir model, rock type identification is an essential task. Recognizing intervals with promising reservoir quality in a heterogeneous reservoir, such as the Pab Formation, using well logs is critical for better exploration, because coring programs are always impractical due to time and cost constraints. Rock types are described by specific log responses, which are ultimately distinguished with the help of electrofacies. The current study uses a cluster analysis technique for the evaluation of reservoir rock types in the identified sand units. K-means cluster analysis is employed to define electrofacies, which are ultimately classified into four rock types on the basis of reservoir quality, from bad to excellent. Rock typing using cluster analysis has been done for four wells, and a correlation has been made to depict changes in electrofacies. From well-to-well correlation, it can be inferred that the reservoir quality of the Pab Formation at the lower portion of Zamzama-02 and 05 wells is excellent and is defined by rock type 4. The Zamzama-03 well in the southwestern region, on the other hand, has good to moderate reservoir quality, as demonstrated by dominating rock types 3 and 2, respectively. The applied prediction technique to the studied field provides continuous rock type identification for the entire reservoir. Using this methodology in defining rock type is cost-effective, requires less time in the demarcation of zones of interest, and is more accurate than manual analysis of the heterogeneous and thick Pab Formation. The studied approach is not only useful in the exploitation of the heterogeneous Pab Formation but also can be applied to other heterogeneous sandstone reservoirs elsewhere.