Modelling and optimisation of a hydrocyclone classifier as a pre-treatment stage in fine coal beneficiation

Coal goes through several types of beneficiation processes that seek to retrieve nearly as much coal as possible in all size fractions. When approaching the finest size of extraction, clean coal extraction becomes limited, hence, there is an increase in fine coal dumps across the top 10 coal-producing countries. Since several extraction processes depend on size, the separation becomes size-limited, i.e., being able to work optimally in a certain size range, be it fines or ultra-fines. The application of hydraulic classifiers and mechanical sieve screens becomes limited when approaching the finest size of extraction. The hydrocyclone classifier has been widely employed to pre-treat the feed in terms of size before fine coal cleaning using other advanced technologies due to its higher centrifugal force. While hydrocyclones serve a crucial function, several factors contribute to the complexity of hydrocyclones in mineral processing plants. In this study, the pre-treatment of coal with an ash of 34.0% was investigated. Several factors of the hydrocyclone classifier were considered, i.e., particle size, spigot size, inlet pressure and the % solids. The modelling and optimisation were achieved using the Taguchi design of experiments, which was an L25 orthogonal array that was developed using Minitab statistical software. The optimal operating parameters for the hydrocyclone were 10% solids w/w, a 20 mm spigot, 770 um F80, and an inlet pressure of 50 kPa. The best outcomes demonstrated a notable improvement in the quality of the coal, with a 30.3% reduction in the underflow ash. The coal yield was evaluated to be 79.2%. The study demonstrated the capability of the hydrocyclone to pre-concentrate fines.