Numerical modelling of two stoping methods in two Indian mines using degradation of c and mobilization of φ based on Q-parameters

Two Indian mines are the subject of a comparative study of a strain-softened Hoek–Brown and FLAC 3D modelling, and a novel ‘c then tan φ’ strain-softening-strain-mobilization approach, using Q-system based input data. This approach is also used with FLAC 3D, using identical stope geometries. The parameters CC and FC, denoting the cohesive component and frictional component of shear strength, are extracted directly from the Q-logging and knowledge of UCS, and are the source of the peak values. Measured deformations, or the strains recorded over the total length of pre-mining installed MPBX, are compared and effectively calibrate the models, in view of the very similar deformations obtained from empirical formulations based on Q using the competence factor approach, as in SRF. The ‘c then tan φ’ approach appears to give the most realistic match to observations in the mines, including the modelling of a shear band within the back or roof of a stope, rather than at the surface of the stope. The Q-based approach also uses a depth-dependent modulus, and this is perhaps the reason why the strain-softened Hoek–Brown model, without this stiffening with depth, shows ‘global failure’ in a second mine having a wider range of depths within one model, and many openings, since modulus is not increased in standard-method approaches.

► Calibration of FLAC 3D models to measured and empirically calculated displacements.
► Input parameters derived from extended Q-system.
► Use of c then tan φ approach (cohesion degradation and friction mobilization).
► Comparison of suggested approach with Hoek-Brown rock mass parameter derivation.
► The suggested approach gives more realistic results and matches displacements.