Numerical investigation of fluid flow in intersecting fractures based on photogrammetry

Summary

Abstract

Fluid flow through fractured rock networks is controlled by several factors, including hydraulic pressure gradients, fracture density, roughness, aperture, connectivity, and intersections. Characterizing hydraulic behaviour at fracture intersections is essential for understanding flow in rough fractured rock masses, especially for nuclear waste disposal. In this study, a block of Finnish Kuru granite (10 cm × 10 cm × 10 cm) was prepared with two perpendicular and intersecting fractures. Photogrammetry was applied to reconstruct three-dimensional (3D) point clouds of the sample. To scale 3D models obtained by photogrammetry, markers with predefined distances were used, and accuracy was assessed by using Root Mean Squared Error (RMSE). Subsequently, roughness and physical aperture of fractures were quantified. The 3D point clouds were converted into a mesh domain in MeshInspector and imported into COMSOL Multiphysics, where fluid flow was simulated by solving the Navier–Stokes equations. Simulations were conducted under hydraulic pressure gradients ranging from 10 to 180 kPa/m, in increments of 10 kPa/m. The influence of intersections on flow for two inlets and two outlets was investigated. The Forchheimer equation was used to analyze the numerical results. The photogrammetric reconstruction accuracy yielded an RMSE of 15 μm. Physical apertures for the two inlets were 190 and 192 μm, while for Outlets 1 and 2, they were 177 and 192 μm, respectively. 3D Roughness values were similar, ranging from 8.7 to 9.3. Numerical results showed nonlinear flow behavior at the outlets. The Forchheimer fitting was used to calculate hydraulic apertures and nonlinear parameters. A higher flow rate with a larger physical aperture, demonstrating the role of geometry in flow distribution. This research shows that fracture intersections influence fluid flow behavior, leading to discharge changes based on geometrical properties, such as physical aperture. These factors must be considered in fracture network flow analysis.