The operation of flotation plants has been improved over the years by use of fundamentally based design techniques, a better understanding of reagent chemistry, online metal analysis and the use of digital control systems. However, optimal operation of flotation banks is not consistently achieved using manual operation. The reason for this is that the system is a highly interacting one, with changes in air addition, cell interface level and reagent dosages all being available, without it necessarily being clear as to which change should be made for a particular situation.
This situation is made more complex by the fact that the feed to a flotation bank can vary in flowrate, composition and density over a fairly short period of time. It is not possible for a human operator to reject these disturbances, given the extent of the operation that is typically being monitored.
Model predictive control (MPC) is a control and optimisation technique, developed in the oil refining industry, that aims to manage just such interacting multiple input multiple output (MIMO) systems. This technology makes use of linear dynamic models of the plant, that are experimentally derived. These models are then used in a receding horizon strategy to determine an optimal trajectory of the inputs to the system.
MPC has been applied to four areas of the lead zinc flotation unit at Mount Isa mines. A total of 31 cells are being controlled by the system, which varies air to the cells, levels and reagent addition on a minute-by minute basis. The MPC controls concentrate and tail grades, reagent dosages, as well as recoveries to ranges set by the plant metallurgists.
The development of the MPCs is discussed, together with the benefits realised. The key factors in insuring an ongoing return on investment are described.
Price, A, Okle, D and Brooks, K, 2018. Glencore Mount Isa lead zinc flotation circuit – experiences from the implementation of model predictive control, in Proceedings 14th AusIMM Mill Operators' Conference 2018, pp 101–110 (The Australasian Institute of Mining and Metallurgy: Melbourne).