2D simulation model of mineral particles yielded by discriminatory size reduction

One of the fundamental problems in mineral processing is to predict how ores are comminuted and how minerals are distributed in the particles. In this paper, image analysis and data processing techniques are applied to support the development, calibration and validation of 2D simulation model of particles generation by size reduction, where minerals exhibit discriminatory behaviour. Based on images of ore textures, a simulation model of fracture lines propagation was developed. A modification of Nelder and Mead optimisation method was used to find out the model’s parameter values that minimise an objective function defined as the squared deviation between observed and simulated particle areal distributions taking into account the variation of the average grade of the target mineral with particle size.