Predictive control of the bubble size distribution in a two-phase pilot flotation column

• Dynamic modelling of the bubble size distribution in a pilot flotation column.
• Constrained nonlinear predictive control of bubble size distribution.
• Mean and standard deviation of the lognormal distribution are controlled.

Among the gas dispersion properties affecting flotation column behavior, the bubble size distribution (BSD) is one of the most important. Its control could positively contribute to optimize the metallurgical performance. Experimental tests are carried out in a laboratory flotation column working with air and water. The objective is to regulate the BSD to a desired distribution set-point. BSD is measured in real-time using an image analysis method and a dynamic non-linear model (Wiener) for BSD, based on a log-normal distribution, is identified. Finally, a constrained model predictive controller is designed to control the BSD. The proposed approach leads to good control results, thus confirming the possibility to use this strategy to optimize the valuable mineral recovery by adequately selecting the BSD for a given particle size distribution.