Research on multi-scale digital core construction of carbonate rock with fractures and caves characteristics: a case study of Dengying formation carbonate rocks in Sichuan Basin

Carbonate reservoirs characterized by fractures and caves exhibit a complex spatial distribution and significant heterogeneity. The establishment of precise and reliable three-dimensional geological models is imperative to elucidate the influence mechanism of fracture and cave parameters on macroscopic petrophysical properties from a micro scale. This is the foundation and prerequisite for the efficient development of such reservoirs. In this paper, dual-scale CT core scanning data is utilized to adjust the segmentation threshold of high-resolution (small-scale) cores to the segmentation of low-resolution (large-scale) cores. This approach enables the comprehensive characterization of multi-scale porosity in large-scale cores and the realization of multi-scale digital core fusion. A 3D fracture digital core model was generated using a Slice-GAN neural network model and SEM images. A well-developed cave was extracted from low-resolution CT data and used as a cave digital core. Through model superposition, a multi-scale digital core containing fractures and caves was constructed. The reliability of the constructed digital core was verified based on four parameters: aperture distribution, coordination number, porosity, and resistivity. The research outcomes establish a foundation for subsequent simulations, which aim to assess the resistivity response of the digital core under varied fracture size, angle, and cave size conditions. This provides a technical foundation for advancing the fundamental theoretical research of carbonate rocks.