Pore-network extraction algorithm for shale accounting for geometry-effect

Abstract

Extracted pore-network modeling is an efficient and reliable way to provide a platform for mathematical simulation of fluid flow in porous media and predicting the transport properties. However, the existing algorithms for pore-network extraction have deficiencies in capturing the unique features of shales, including the geometry information and the complex connections between adjacent pore bodies. The newly developed method, based on the maximal-ball algorithm, proposes a novel and enhanced algorithm for the classification of pore bodies and throats. The extremely small pores, composed by only several voxels, can be captured by the new method. We capture the pore-throat geometry by characterizing the cross section of each maximum inscribed sphere connecting the two adjacent pore bodies; the pore-body geometry is characterized using the locally largest maximum inscribed sphere and the overlapping ones. The novel algorithm takes into account the complex connection between the adjacent pore bodies in determining the pore-throat size, where the flow capacity is maintained. A Marcellus shale core sample is used to validate our new method. The core sample is scanned using scanning electron microscope with the resolution of 4 nm, and the three-dimensional reconstruction of the porous media is built. We extract the pore-network model based on the three-dimensional reconstruction and analyze the statistic of the properties, like the distribution of pore size and geometry factor. The pore structures are quantitatively compared with the structures extracted by the previous maximal-ball method. The results show that our new method can capture more extremely small pores than the previous method and the classification of pore throats and bodies is more reasonable. So, the newly developed method is more reliable to extract physically realistic pore networks.