Block caving is the most cost effective and productive underground mining method for massive ore bodies. The feasibility, design and continuous production of the block caving method depend on key parameters related to the geological and geotechnical characteristics of the ore body. The forecasting of the size distribution of the fragmented rock is one of these parameters. The fragmentation distribution depends mainly on rock strength and defects, in-situ and induced stresses and the degree of fracturing. In addition to the challenge of characterizing each of these parameters, the size fractions of interest are located at the extremes of the distribution, where data are scarcer and statistics are more affected by random variations. Application of discrete fracture network to block caving fragmentation: a hybrid approach. Thus, average and central tendencies of the distribution are not of much practical use.
One of the most often used tools to assess fragmentation for block caving projects and mines is the Block Caving Fragmentation (BCF) software, created by G Esterhuizen. In the new DFN-BCF hybrid approach, BCF’s primary blocks generator is replaced with a Discrete Fracture Network (DFN) model of the rock mass, making use of the full geometry of the relevant discontinuities that drive the fragmentation process. The final product of the assessment corresponds to a weighted fragmentation curve, representing the complete block to be caved, and its corresponding range of variation. The DFN realisations ar generated in the software FracMan®. The primary blocks from the DFN realisations are then processed using the BCF algorithm, then coded into a Python program to compute the splitting of blocks and produce the secondary fragmentation curve. A novel graphical output has been added to the assessment, consisting of a scaled graphical representation of the actual size distribution using 3D spheres.
Since its creation in mid-2020, the DFN-BCF hybrid approach has been successfully applied to two world-class orebodies. The results from the early stages of operation from one of these has shown a good agreement between the forecasted fragmentation and the actual block sizes at the drawpoints.
