Fractal Analysis Of Ct Images Of Tight Sandstone With Anisotropy And Permeability Prediction

For developing unconventional oil and gas resources, it is necessary to quantify the heterogeneity embodied in both pore-size distribution (PSD) and pore spatial distribution, on the basis of which the relationship between the seepage characteristics and the microstructure of heterogeneous porous media could then be clarified. In this work, micro computed tomography (mu CT) and image processing technology were applied to characterise matrix pore structure in tight sandstone. Fractal theory was used to investigate the image-based heterogeneity of tight porous media. On the basis of the verification that the PSD of the tight rock is statistically self-similar at a certain scale, we investigated the relationships between porosity and fractal dimension, and between 2D and 3D fractal dimensions, respectively. The results showed that the pore structure characteristic parameters with anisotropy, i. e., the ratio of the minimum pore size to the maximum pore size lambda max/lambda min and a proportionality constant CKT, could be obtained by liner fitting fractal dimension and porosity. Moreover, the 2D PSD characteristics of the 'standard cross section' were consistent with the 3D PSD characteristics of pore network. A fractal permeability model was established by treating the connected pores in the matrix as a bundle of conical micropipes. The permeability values predicted by the proposed model agreed well with those predicted by finite volume method simulation under the condition where irreducible pores were removed. A sensitivity analysis indicated that increasing the heterogeneity of the PSD and pore spatial distribution reduced the flow capacity of the matrix, particularly for low-porosity porous media.