An efficient 3D cell-based discrete fracture-matrix flow model for digitally captured fracture networks

Abstract

Complex hydraulic fracture networks are critical for enhancing permeability in unconventional reservoirs and mining industries. However, accurately simulating the fluid flow in realistic fracture networks (compared to the statistical fracture networks) is still challenging due to the fracture complexity and computational burden. This work proposes a simple yet efficient numerical framework for the flow simulation in fractured porous media obtained by 3D high-resolution images, aiming at both computational accuracy and efficiency. The fractured rock with complex fracture geometries is numerically constructed with a cell-based discrete fracture-matrix model (DFM) having implicit fracture apertures. The flow in the complex fractured porous media (including matrix flow, fracture flow, as well as exchange flow) is simulated with a pipe-based cell-centered finite volume method. The performance of this model is validated against analytical/numerical solutions. Then a lab-scale true triaxial hydraulically fractured shale sample is reconstructed, and the fluid flow in this realistic fracture network is simulated. Results suggest that the proposed method achieves a good balance between computational efficiency and accuracy. The complex fracture networks control the fluid flow process, and the opened natural fractures behave as primary fluid pathways. Heterogeneous and anisotropic features of fluid flow are well captured with the present model.

Forecasting residual mechanical properties of hybrid fibre-reinforced self-compacting concrete (HFR-SCC) exposed to elevated temperatures

“…are already common in the field of civil engineering. These techniques and several other algorithms are frequently being utilized to estimate different properties of concrete composites, soil compaction parameters, slope failure susceptibility etc [ [28] , [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] , [39] , [40] ]. This research is attributed to utilization of a special ML algorithm called GEP to predict residual cs and fs of SCC containing a mix of steel, polypropylene and PVA fibres.…”