Determination of REV size and equivalent permeability coefficient of fractured rock masses from the perspective of fracture network connectivity

• Published in: Computers and geotechnics 2024-09, Vol.173, p.106590, Article 106590
• Main Authors: Wei, Li, Zhechao, Wang, Liping, Qiao
• Format: Article
• Language: English
• Subjects: Connectivity; Fracture network; Permeability; Representative elementary volume (REV); Size effect
• ISSN: 0266-352X; 1873-7633
• DOI: 10.1016/j.compgeo.2024.106590

Hydraulic properties of fracture rock masses exhibit a strong correlation with the size and connectivity of the fracture network. Based on the connectivity analysis of the fracture network, the size effect on the permeability was investigated. A method for analyzing the connectivity of the fracture network was proposed based on the graph theory and the connectivity of the rock fracture network in unidirectional and radial flow configurations were analyzed using the method. The percolation parameter was used to describe the connectivity of the fracture network and the critical percolation parameters for connected and disconnected fracture networks in two configurations were 0.353 and 0.230. The size effect on the permeability of the fracture network was investigated and the influence of the fracture density and size on the representative elementary volume (REV) size for permeability and permeability coefficient were obtained. The results showed that the REV size decreased but the equivalent permeability coefficient increased with the fracture density and size. Based on the above results, the relation between the REV size for permeability and the fracture geometric parameters was established and the REV size with different percolation parameters could be determined easily according to the relation. Considering the connectivity of discrete fracture network (DFN), the calculating expressions for the permeability coefficient tensor in unidirectional flow configuration and radial permeability coefficient in radial flow configuration were proposed. The groundwater inflow into caverns was predicated using these expressions and the predicted results were very close to the measured results in the field.