Rock Typing and Reservoir Quality Analysis of the Abu Madi Reservoir: Distribution Prediction Using Artificial Neural Networks in the West El Manzala Area, Onshore Nile Delta, Egypt

Abstract

AbstractIn the Nile Delta gas province of Egypt, the Abu Madi Formation is the most promising sandstone gas reservoir. This study aimed to investigate reservoir rock typing and quality by integrating petrophysical and petrographical data, including well logs, image logs, and cores. Furthermore, it sought to predict the permeability and reservoir quality of uncored wells by integrating artificial neural network technique with core analysis data and evaluate the effectiveness of this approach as an exploration tool in the West El Manzala area. The core petrography revealed the presence of microfacies consisting of arenites and wackes. The measured porosity, permeability, and pore sizes obtained from the cores, along with the parameters of the reservoir quality index, normalized porosity, and flow zone indicator, indicated that the Abu Madi reservoirs could be subdivided into three categories based on reservoir quality. High reservoir quality (RT-I) is characterized by megapores within the hydraulic flow unit (HFU-1) associated with bioturbated coarse to gravelly sandstone facies. Moderate reservoir quality (RT-II) is characterized by macropores within the hydraulic flow unit (HFU)-II associated with massive coarse to gravelly sandstone facies. Poor reservoir quality (RT-III) was characterized by mesopores within the HFU-III hydraulic flow unit related to laminated silty mudstone facies. The findings of this study demonstrate that the combination of reservoir rock typing and artificial neural networks is an extremely successful method for petroleum exploration in the West El Manzala region.