Fracture modeling of the hydrocarbon reservoir using geostatistical and neural network methods in the SW Iran Oilfield

Abstract

Fracture characterization and modeling are essential for hydrocarbon exploration and enhanced production. To model the fracture network in the Asmari reservoir of the Rag-e-Sefid Oilfield (SW Iran), this research characterizes fracture intensity using well, fracture driver, and fracture controller data. First, these data are analyzed to estimate fracture intensity. Then, fracture intensity is modeled using geostatistical methods. The geostatistical outputs are compared and calibrated based on the structural setting of the study area and the fracture indicator. Finally, selected fracture intensity data are integrated into a single model using an artificial neural network, resulting in a comprehensive fracture intensity model for the Asmari reservoir of the Rag-e-Sefid Oilfield. The results show that fracture intensity increases near the Rag-e-Sefid and Nourooz-Hendijan-Izeh Faults and in the fold forelimb and crest. The highest fracture intensity in the Asmari reservoir is observed at the intersection of structures with the N-S Arabian trend and the NW-SE Zagros trend, where the fold axis has rotated. Generally, the northwestern part of the Rag-e-Sefid anticline has higher fracture intensity than the southeastern part. The high fracture intensity in the northwest part of the Rag-e-Sefid Oilfield is related to inversion tectonics, multi-stage reactivation along pre-existing basement structures, and an older deformation history in this area compared to its southeastern part. The Asmari reservoir in the NW part of the Rag-e-Sefid anticline contains a greater share of oil and gas in its hydrocarbon traps than the SE part. Moreover, the results of this study indicate that the simultaneous use of different data and the integration of geostatistical and artificial neural network methods can effectively predict fracture distribution in hydrocarbon reservoirs and be used as a suitable technique for fracture modeling in natural oil and gas fields. This research suggests that artificial intelligence and quantum computing techniques provide efficient solutions for characterizing and modeling the entire scale of geological fractures in hydrocarbon reservoirs.

Keywords: Fracture modeling, Geostatistical and neural network methods, Asmari reservoir, Rag-e-Sefid Oilfield, SW Iran