Abstract
A novelty-designed wide-neck classifier (WNC) was proposed to enhance the passing ability and classification efficiency of fine particles. Using computational fluid dynamics (CFD), we studied the flow field and velocity distribution in the newly designed WNC. The velocity of the fluid gradually decreased from the wall to the center and from the cylinder to the cone, which facilitates particle classification and thickening. The Reynolds number (Re) and turbulent intensity (I) inside the WNC were discussed. The turbulent intensity increased with increasing feed velocity and overflow outlet diameter and decreased with increasing feed concentration and spigot diameter. The classification of coal slurry was performed to analyze the performance of WNC. The classification efficiency increased with increasing feed velocity but decreased as the feed concentration, spigot diameter, and overflow outlet diameter increased. The predictive models for classification efficiency influenced by the operational and structural parameters were constructed at high correlation coefficients, and the average error of these models was analyzed at 0.28%. Our results can provide valuable insights into the development of mineral classification.