Resumen

ROBERTS, D.P.; HOFMANN, G.F.; NEL, R.  y  SCHEEPERS, L.J.. Modelling of generic excavation sequences for bore-and-fill mining. J. S. Afr. Inst. Min. Metall. [online]. 2019, vol.119, n.10, pp.833-846. ISSN 2411-9717.  http://dx.doi.org/10.17159/2411-9717/687/2019.

Bore-and-fill mining refers to the mechanical excavation of ore through a series of holes bored on-reef and the subsequent filling of these holes with a high-stiffness material. A calibrated numerical model was used to evaluate the influences of various design parameters on bore-and-fill operations for the Carbon Leader Reef. A set of metrics was prepared and a small-span generic model used to evaluate the influence of boring sequence, fill properties, stress regime, and hole location. The modelling indicated that most metrics stabilize when the number of hole diameters skipped between bored holes is three or greater. Increasing the skip beyond three reduces the potential for regional failure but may increase the potential for local failure. The recommended design is to skip three or more holes. Variation of the fill properties showed that the Young's modulus has a nonlinear effect on design metrics. Closure volume was found to increase sharply when the fill stiffness is decreased below 20 to 25 GPa. The minimum suggested fill stiffness is therefore 25 GPa, where the fill performance will be within 5% of the reference case (filling with intact rock mass). The material cohesion had a linear effect on the performance metrics. A laminated model resulted in slightly increased closure and revealed that the worst-case scenario for hole position is when all the holes are drilled mostly in the weaker reef layer. Altering the stress field to match recent stress measurements did not significantly affect the results. Increasing the driving stress by 28% had a proportionally greater effect on all metrics (25 to 45%). It was noted that implementing bore-and-fill at greater depths will require re-evaluation of the fill and sequence design. Increasing the span by 14.2 m to 34.8 m (and adding pillars and reef raises) resulted in an 18% increase in maximum closure and a 34% increase in the closure volume increment. The damage and general behaviour was very similar to the smaller span models. The difference in metrics for pillars sizes of 7.6 and 6.3 m was not significant. Use of the calibrated model allowed for optimization of sequence and fill design parameters. The limits of the model were also explored and it was found that a larger fine-mesh zone would be needed for models at greater depth. Laser scans of the reef drives and raises indicated rock mass deformations from 1.3 to 3.2 times the modelled closure. Though these were greater than the modelled values, it was noted that the in situ deformation also reflects time-dependent behaviour and the response of the rock mass to seismic events. The model results were assessed in terms of the observed seismic response during mining of the pilot site. It was found that the seismic response was limited and was associated with the boring of holes and limited fracturing within existing abutments. This supports the model results, which indicated limited closure volume and very little potential for the formation of seismogenic shear fractures. An elastic boundary element model was calibrated using the model outcomes, and it is shown how this model can be used to assess layouts involving bore-and-fill mining in terms of closure volume and other rock engineering design parameters.

Palabras clave : bore-and-fill mining; numerical model; rock mass deformation; seismicity.

        · texto en Inglés     · Inglés ( pdf )