Three-dimensional discrete network modeling of structural fractures based on the geometric restoration of structure surface: Methodology and its application

摘要

Structural fractures are widely developed in the oil and gas fractured reservoirs. However, it is difficult to characterize structural fractures accurately in three-dimensional (3D) space by using: core description method; logging interpretation method; tectonic curvature method; stochastic simulation method; or seismic description method. Based on the theory of fault-related folds and stochastic simulation, a method for constructing 3D structural fracture models relying on the geometric restoration of the structure surface was developed. The structural surface geometrical restoration technique was applied to establish the structural fracture development index model. The core description results were utilized to construct the structural fracture group mode and the fracture intensity model for each fracture group. In addition, the fracture development intensity model was developed and used as the main input to construct the 3D discrete network structural fracture model based on the spatial restriction to the structural fracture development index model. This model achieves 3D visualized modeling for each group of structural fractures and their attribute characteristics. A case study was performed by applying this proposed method to a Jiaoshiba shale gas field in the braided fold belt of eastern Sichuan Basin, where a discrete network model of structural fractures with high and low angle was established. This discrete network model reconstructed the distribution, development scale, extension direction and cutting relationship of structural fractures in 3D space. The modeling results not only provide a discrete structural fractures network model for simulating the production of the Jiaoshiba shale gas field, but also provide a new method for developing a discrete structural fractures network model for sandstone reservoir, carbonate reservoir and shale reservoir.