3D modelling of soil-rock mixtures considering the morphology and fracture characteristics of breakable blocks

Failure mechanisms of a soil-rock mixture (S-RM) can be efficiently investigated by the discrete element method (DEM). This paper proposes a stochastic approach for 3D DEM modelling of S-RM samples accounting for morphological features and internal fractures of blocks and their potential breakage. The research refers to the case-history of an artificial S-RM filling slope, mainly containing highly-weathered shale blocks, constructed at the ±500 kV electronic converter station in the Funing County, Yunnan Province, China. The 3D morphological features of the blocks and the characteristics of their internal fractures were obtained by CT technology and image processing. A stochastic method based on harmonic series was developed starting from the real blocks, allowing the generation of random block clusters with characteristics similar to the original ones. Another stochastic approach was implemented for the creation of the internal fractures, simulated as voids in the block clusters, following the characteristic of the real fractures. Finally, the procedure was applied for the definition of a 3D S-RM DEM model with 40% block proportion, whose meso-parameters were determined for simulating direct shear tests. These latters were also useful to explore the mechanical response of the sample at the mesoscale, including the formation and development of the localization band. The numerical results showed that the generated random S-RM DEM model well reproduced the experimental behaviour of S-RM samples with breakable blocks. Also, they highlighted the importance of modelling the block breakage and internal fractures; in fact, companion simulations with unbreakable blocks and breakable blocks without fractures were all characterised by increased shear strength with higher friction angle but reduced cohesion.