Pore network extraction for fractured porous media

•Development of a pore-network extraction method for rocks with intersecting fractures embedded in a porous matrix.•Implementation of an efficient shrinking algorithm for simultaneous extraction of medial axes and medial surfaces for arbitrary complex geometries.•Conversion of the skeleton of axes a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Advances in water resources 2017-09, Vol.107, p.280-289
Hauptverfasser: Jiang, Z., van Dijke, M.I.J., Geiger, S., Ma, J., Couples, G.D., Li, X.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•Development of a pore-network extraction method for rocks with intersecting fractures embedded in a porous matrix.•Implementation of an efficient shrinking algorithm for simultaneous extraction of medial axes and medial surfaces for arbitrary complex geometries.•Conversion of the skeleton of axes and surfaces into an integrated pore-network model suitable for single- and two-phase flow simulation in fractured rocks. Although flow through fractured rocks involves many different length-scales, it is crucial for the prediction of continuum-scale single- and multi-phase flow functions to understand, at the pore-scale, the interaction between the rock matrix and fractures. Here we present a pore-network extraction method in which the pore diameters and fracture apertures are of similar size. The method involves a shrinking algorithm to extract a hybrid skeleton of medial axes and surfaces, and it includes a workflow to convert the medial surfaces of fractures into dense networks of virtual medial axes, allowing generation of an integrated pore-network for the entire pore space. Appropriate single- and two-phase flow properties are assigned to network elements representing the fractures. We validate the method via comparisons between pore network flow simulations and an analytical solution, direct flow simulations and experimental observations. The network calculations are several orders of magnitude faster than the direct simulations.
ISSN:0309-1708
1872-9657
DOI:10.1016/j.advwatres.2017.06.025