Improving the accuracy of fracture modeling in carbonate reservoirs X-field in SW of Iran

Fracture modeling is one of the most important steps in the study of fractured reservoirs. Due to the high cost of imaging logs and their absence in most wells of the study area, it is often attempted to use other available data to detect fractures. This paper attempts to investigate the relationship between the lithology and fractures of rocks. For this purpose, the Image, Neutron, Density, Litho-density, and NGS logs have used to simulate the lithology. Based on this feature, the studied area was divided into six homogeneity part, and the fracture probability was determined in each section to improve the accuracy of fracture modeling. Recently, an intelligent method has been proven as an efficient tool for modeling complex and non-linear phenomena. In this paper, neural network methods has been used in fracture modeling. The results show that the division of the field based on lithological studies will improves the accuracy of fracture modeling in the studied area up to 7 percent without increasing the cost of image logging.