DFN Variability Analysis Through Voxelization

A fracture network model provides a wealth of information concerning the characteristics of a rock mass, including fracture intensity, block size, and connectivity. This paper identifies methods to utilize information derived from DFN models. The fracture intensity of a DFN can be analysed for individual or grouped subvolumes of the model. A DFN is discretized into volumetric elements (voxels). Stochastic DFN models measuring 25 × 25 × 25 m³ are modeled with identical attributes. Fractures are modeled as non-planar and non-persistent, and orientations varied to generate a variety of block shapes and sizes. The voxels used for metrics analysis range in size from 1 m³ to 1 953 m³, representing from 2³ to 25³ voxels. Fracture intensity is measured as a P₃₂ value for all elements. The coefficient of variation (CV%) of fracture intensity is calculated for each voxelization to generate a curve characteristic of a particular DFN model. An estimate of the in situ block size distribution (IBSD) is made by considering the number of unfractured voxels (null blocks). The fracture intensity, CV% of fracture intensity and IBSD values derived from these DFNs can be used as input to a variety of secondary models.