Premium
Pore network extraction from pore space images of various porous media systems
Author(s) -
Yi Zhixing,
Lin Mian,
Jiang Wenbin,
Zhang Zhaobin,
Li Haishan,
Gao Jian
Publication year - 2017
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1002/2016wr019272
Subject(s) - characterisation of pore space in soil , porous medium , lattice boltzmann methods , medial axis , porosity , permeability (electromagnetism) , network model , materials science , relative permeability , lattice (music) , biological system , parameterized complexity , computer science , algorithm , geology , mechanics , geotechnical engineering , artificial intelligence , physics , chemistry , biochemistry , membrane , acoustics , biology
Pore network extraction, which is defined as the transformation from irregular pore space to a simplified network in the form of pores connected by throats, is significant to microstructure analysis and network modeling. A physically realistic pore network is not only a representation of the pore space in the sense of topology and morphology, but also a good tool for predicting transport properties accurately. We present a method to extract pore network by employing the centrally located medial axis to guide the construction of maximal‐balls‐like skeleton where the pores and throats are defined and parameterized. To validate our method, various rock samples including sand pack, sandstones, and carbonates were used to extract pore networks. The pore structures were compared quantitatively with the structures extracted by medial axis method or maximal ball method. The predicted absolute permeability and formation factor were verified against the theoretical solutions obtained by lattice Boltzmann method and finite volume method, respectively. The two‐phase flow was simulated through the networks extracted from homogeneous sandstones, and the generated relative permeability curves were compared with the data obtained from experimental method and other numerical models. The results show that the accuracy of our network is higher than that of other networks for predicting transport properties, so the presented method is more reliable for extracting physically realistic pore network.