Model Predictive Control of a Zinc Flotation Bank Using Online X-ray Fluorescence Analysers

Abstract

In the metals and mining industry, flotation plays a vital role in selectively concentrating and separating valuable minerals from gangue minerals in a process slurry. The control of flotation circuits has proved difficult in the past, with large variations in reagent dosages, concentrate grade and the recovery of the valuable minerals being experienced. At Glencore’s Mt Isa site, the control issues have been addressed using model predictive control (MPC). The MPC makes use of online grade analysers for precise control and an optimising controller engine for driving the process to an optimal state. The use of the analysers has introduced challenges associated with their relatively slow frequency, and the need to employ sensor validation for the measurements.