Discrete fracture modeling of deep tight sandstone reservoir based on convergent multi-information—a case study of KX gas field in Tarim Basin

With low porosity and permeability, heavy heterogeneity, as well as frequent interlayers, fracture networks act as important flow channels for tight sandstone reservoir and influence the regular oil and gas production in the development stage. With this mind, new and effective modeling and simulation methods are needed to be developed to guide discrete fracture network modeling (DFN) which integrate convergent multi-source information including geology, well-logging, seismic, CT, and dynamic data at different scales. In this paper, we present an entropy weight method fusing a wide variety of geological information to build precise intensity body of different scale fracture networks and use both deterministic modeling and stochastic modeling approaches to establish 3-D simulations of fracture reservoir facies in a tight sandstone reservoir. Finally, a DFN model and a fracture property model are built by means of parameter field equivalence. Since various factors considered, different influencing parameters are taken into account in this approach. It leads to an explicit and multi-scale modeling of fracture, which provides a valid workflow to increase confidence in predicting the generation of fractures and their spatial distribution of deep tight sandstone reservoirs.