Multiscale pore/throat characterization in tight sandstone formation with multi-threshold image segmentation algorithm
Year: 2025
Type: article
Abstract: Introduction Pore space in tight sandstone formation is very complex with micro-scale and nano-scale pores/throats, the multi-scale characteristics needs to be considered for the construction of microscopic pore model accurately. Methods In this paper, micro-CT is used to obtain the representative 3D micro-scale gray image, and nano-CT is used to obtain the representative 3D nano-scale gray image based on the representative nano subsample. Then, the multi-threshold image segmentation algorithm is introduced to segment the micro-scale and nano-scale gray image respectively. Three distinct phases are identified in microscale images: (1) micro pore/throat phase, (2) micro-particle phase, and (3) a transitional matrix phase exhibiting intermediate grayscale values. This matrix phase represents unresolved nano-porosity and nano-particles beyond micro-CT resolution limits. Conversely, nanoscale images reveals two discrete phases: well-defined nanopore/throat phase and nano-particle phase. The micro-pore/throat and nano-pore/throat digital rocks are constructed respectively, and the corresponding pore network model in tight sandstone are extracted, the porosity and pore/throat size distribution at micro scale and nano scale is calculated respectively. Results and discussion Moreover, the nano pore/throat characteristics in nano-pore/throat digital rock is upscaled into the matrix phase of micro-pore/throat digital rock, which could calculate the total porosity and multiscale pore/throat size distribution in tight sandstone formation accurately. The digital rock analysis results are verified correctly with the comparison of the lab test results, which provides a fundamental platform for multiscale pore/throat characterization and formation evaluation in tight sandstone formation with important academic and application value.
Source: Frontiers in Earth Science
Authors: Hu Rongrong, Han Jiangchen, Wang Chenchen, Zhu Jian, Dong Hu
Language: English
Subfield: Ocean Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W4410035728