IMPROVED TRANSFORMATION TECHNIQUE FOR THE PREDICTION OF LIBERATION BY A RANDOM FRACTURE MODEL

King's Random Fracture Model (RFM) was implemented using the exponential model for the distribution of linear intercepts through both continuous and dispersed phases and the Finlayson function for the distribution of intercept lengths through particles. Accurately measured distribution of intercept lengths through particles and linear grade distributions were compared to those predicted by the model, and very good agreement was found. The inversion of the transformation equation, which describes the relationship between the one-, two- and three-dimensional information regarding the composition of mineral particles was implemented using the constrained Rosenbrock Hillclimb Procedure with an appropriate objective function. Volumetric grade distributions determined from depth profile measurement were compared with the results of inversion from either linear or areal grade distributions, and again very good agreement was found. Assuming both procedures were correct, the liberation spectrum of a copper ore containing 3.8 vol.% CuFeS2 was determined from very simple image analysis measurements using the RFM coupled with the implemented solution for the transformation equation.