Modern grinding and flotation circuits operate under a closed-loop, data-driven control system to ensure the stability of ore throughputs, particle size, and recoveries even with varying input feeds. These systems usually employ conventional instrumentation and control techniques along with intelligent controls, considering that the quality of grinding significantly impacts flotation outcomes.
Grinding control involves maintaining a specified target particle size while optimizing throughputs and preventing overloading. In this process, feed rates, sump levels, cyclone pressure, circulating loads, and discharging densities are usually controlled, as they impact residence time and classification processes. The current trend is towards implementing model predictive control, soft sensors, and adaptive controls in managing the interactions between variables.
The concept of coordinated control is emphasized as opposed to individual loop tuning. Investigations into grinding process optimization indicate that the optimal particle size does not remain constant; rather, it needs to vary to enable flotation recovery efficiency and profitability of the entire process circuit, particularly when ore hardness or liberation varies.
Flotation process control entails the maintenance of steady-state pulp chemistry and hydrodynamics. Air flow rate, pulp level, froth level, chemical feed rate, pH, temperature, and pulp density are examples of manipulated variables, while pulp grade, recovery, tailings quality, and flow balance are typical control variables. As flotation processes exhibit multivariate nonlinearity, there is considerable interest in advanced control techniques like MPC and adaptive decoupling, with one pilot-column study indicating adaptive decoupling control to be competitive with or superior to MPC for some criteria, with reduced computational requirements.
Integration of grinding and flotation into one unified system is today’s leading tendency. Such integration acknowledges the possibility of improving mineral liberation by means of reducing grinding size but also takes into account higher energy expenses and possibly altered nature of flotation feedstock, which results in finding the right balance between the two factors as a best set point.
Integrated control relies on analytical equipment, granulometric estimates, expert system analysis, and supervisory algorithms for altering the grind point according to changes in flotation efficiency and mineral properties.
Therefore, process control methods of greatest importance include stabilization of feed stock, multivariable feedback control, either MPC or adaptive control, real time monitoring, and integrated grinding-flotation optimization.
