Integration of permeability estimation methods using geostatistics and artificial neural network; A case study one of oil fields in the Persian Gulf

Permeability is one of the important parameters in reservoir petrophysical studies, and evaluation of this parameter can be used as a key tool in the oil fields development. This study aim is permeability estimation and modeling of the Upper Surmeh Formation in one of the oil fields in the Persian Gulf. The Surmeh Formation with Jurassic age is considered as one of the most important oil and gas reservoirs in the Persian Gulf basin. In this study, we have used petrophysical well logs and 3D post-stack seismic data in the permeability evaluation process. The structural reservoir model has been prepared using the interpretation of seismic sections and well logs in the reservoir section. This model includes the interpretation of fault surfaces, geocell network and reservoir horizons. The geocell network used in this study used columns and geocells with dimensions of 50 * 50 meters in the X and Y directions. The thickness of the geocellular layers of each reservoir zone is designed to fit that zone in the reservoir section. The values permeability estimation was performed using the artificial neural network with a back-propagation algorithm. The results obtained from the artificial neural network were generalized in the studied reservoir well logs. The correlation coefficients value obtained from permeability estimation values with drilling core data is equal to 88%. Comparison of geostatistics results with permeability value shows that the proposed methods can provide acceptable results for reservoir permeability modeling.