A New Approach to the Grinding Kinetics of Magnetite Ore Based on the Population Balance Model

Abstract

A new approach to batch grinding kinetics was established based on the conventional population balance model, with magnetite as the experimental object. The distribution function commonly used in the population balance model is a sum of two power functions, i.e., Bi1=φ(xi−1x1+1−φ(xi−1x1)β. Based on the new finding that the cumulative mass fraction coarser than the size class of the discharge is consistent with the first-order grinding kinetic, the gi function of the new approach is only a single power function, i.e., =+k1xia, which will greatly reduce the parameter error and make the fit more accurate. The maximum error between simulation calculations and the actual experiment using the two methods did not exceed 1%, indicating that both models can accurately predict the fracture characteristics of magnetite. Because the new approach has fewer derived parameters, it addresses the conventional population balance model’s problems of large computational effort and poor fitting accuracy, making it more applicable to the study of the impact of parameters on the grinding status, with a simpler process and higher accuracy. In addition, this new method is applicable to minerals other than magnetite. Further research is required to verify its applicability to wide size ranges and continuous grinding.