A fracture modeling method for ultra-deep reservoirs based on geologic information fusion: an application to a low porosity sandstone reservoirs in X gas field of a basin in western China

Abstract

The reservoirs of ultra-deep and low-permeability sandstones typically exhibit characteristics of lithological tightness and poor physical properties. Fractures control the oil and gas content as well as the productivity of such reservoirs. However, the distribution of fractures is complex, exhibiting strong heterogeneity. Therefore, a systematic study on reservoir fracture modeling can provide geological foundations for the development of such reservoirs. Due to the considerable burial depth of these reservoirs, conventional methods relying solely on seismic information have limited reliability, and the established discrete network models of fractures are often less dependable. In this paper, taking the X gas reservoir in a basin in western China as an example, we discuss a fracture modeling method based on the integration of geological information to enhance the efficiency and accuracy of fracture modeling. The modeling method primarily involves the use of deterministic methods to obtain large-scale fractures, while random simulation is employed for small and medium-scale fractures. The fracture development control factors and seismic attribute information are integrated using permanence of ratios (PR) model to establish a fracture development probability field model. Subsequently, the geometric parameters of fractures and the fracture density model are used as input parameters to generate a discrete network model of small and medium-scale fractures using a object-based modeling method. Finally, based on the fracture equivalent property model and verified through geological understanding, analysis of production dynamics, and numerical simulation of the gas reservoir, it is demonstrated that the fracture model established using the proposed method aligns with geological understanding and exhibits high reliability.