Model fine tuning for prediction of hydrocyclone performance—An industrial case study

Investigation of overall solid recovery and grade efficiency of hydrocyclones is fundamental to the computation of mass balances and underflow/overflow grade distribution in processes in which these equipments are installed. In this work, we approach these issues by considering two models taken from the literature: the Leith–Licht model, developed to predict the performance of gas cyclones, and the Plitt model which is applicable to both gas and hydro cyclones. The results show that the Leith–Licht and the Plitt models are suitable for crude prediction of hydrocyclone performance when the estimation of model parameters is aided using correlations available in the literature. However, during operation, an improvement of the hydrocyclone performance prediction (grade and overall efficiencies) is possible by fitting the model parameters to process data. In this work, a fine tuning of the Plitt model was conducted and the results showed an interesting application on the design of new hydrocyclone arrangements as well as on defining their modes of operation in an aluminum refinery.