Economic model predictive control for a rougher froth flotation cell using physics-based models

The development of an economic model predictive control (E-MPC) strategy is presented. uses a novel dynamic flotation that incorporates the physics froth phase in cell. was previously calibrated and validated using experimental data. Sensitivity analyses were conducted to select suitable objective function accounted for both process economics variable sensitivities. While ultimate goal rougher cell maximise metallurgical recovery at steady state specified minimum grade, it evident incorporation air dynamics (which can be measured real-time) concentrate grade (calculated through first-principle models) led best results. addition easily online, i.e. recovery, offers great potential improve plant performance existing systems. Furthermore, imposed E-MPC strategy. This acts as constraint allows optimised while ensuring requirements are met. optimisation problem discretised orthogonal collocations, implemented Matlab automatic differentiation via CasADi. Two typical manipulated variables considered: flowrate pulp height setpoints. Based on laboratory-scale data, implementation resulted improvements ranging from +8 +22 % maintaining grade. therefore encouraging explore larger