Explicit incorporation of discrete fractures into pore network models

Chenhui Wang*, Kejian Wu, Gilbert G. Scott, Alfred Akisanya, Quan Gan

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Fractured porous media exist widely in reservoir rocks and fractures play a critical role in fluid flow processes. High resolution direct numerical simulations of fluid flow can provide important insight into pore-scale processes and flow mechanisms in fractured permeable media, however, the tremendous computational costs prevent these methods from being applied into larger scale models. Pore network modelling is an alternative solution to this problem, but currently it can only be used in porous media without fractures. This work is concerned with incorporating discrete fractures into pore network modelling to represent fractured porous media. A new fracture matrix pore network model
(FM-PNM) is developed to efficiently simulate the fluid flow properties in fractures associated with the pore matrix. Discrete fractures are transformed into a fracture pipe network and integrated with the pore matrix that is represented by a network of pore bodies and pore throats. These two networks are coupled together to create a single nested network which is topologically equivalent to the fractured porous medium. This method extends the scope of applications of pore network modelling to fractured porous media. The permeability of the coupled network (FM-PNM) is benchmarked by Lattice Boltzmann simulation for various structures of pore matrix and discrete fracture networks.
Original languageEnglish
Article numbere2021WR031731
Number of pages24
JournalWater Resources Research
Volume58
Issue number12
Early online date15 Dec 2022
DOIs
Publication statusPublished - Dec 2022

Keywords

  • fractured porous media
  • fluid flow simulation
  • fracture matrix pore network moderling

Fingerprint

Dive into the research topics of 'Explicit incorporation of discrete fractures into pore network models'. Together they form a unique fingerprint.

Cite this