IM’s January issue will feature automation, in the mine. But in the processing plant, ABB’s Expert Optimizer has recently been used to automate Boliden’s zinc flotation circuits. The Garpenberg plant in Sweden is benefiting from successful application of Model Predictive Control (MPC) to a complete flotation circuit.

ABB says it successfully applied its award winning Expert Optimizer solution to the zinc flotation circuits at Garpenberg. The solution applied MPC to help to stabilise the overall process, and maximise zinc revenue.

ABB and Boliden have a long track record of successful collaborations in the field of automation technology and agreed to collaborate on a project to apply MPC to a complete flotation circuit. The aims were twofold. Firstly to stabilise the process to compensate for external factors, such as ore quality changes. Secondly, the zinc revenue should be maximised, through maximised recovery and higher concentrate grade.

Successfully optimising the circuit required that three criteria be met. First, the behaviour of the process had to be predicted using a reliable dynamic model of the flotation circuit. Then cost functions and finally operating constraints were incorporated in order to produce a robust MPC application.

In order to achieve certainty with regard to the controller benefits, a measurement campaign was started at the end of 2008 and finished in the middle of 2009. The results showed that concentrator efficiency (directly corresponding to revenue) was at least 1% higher when MPC was used, compared with the existing manual control strategy. In addition, the zinc concentrate grade has been higher and more consistent, while the recovery has remained at the same level.

Henrik Lindvall, Process Engineer at Boliden, Garpenberg, said, “Following the successful testing period, Boliden is continuing to use Expert Optimizer on their Garpenberg zinc circuit and has continued to report good results throughout 2009”.

Expert Optimizer, part of ABB’s cpmPlus suite of applications, is the recommended tool, by ABB, for implementation of advanced process control in general, and MPC in particular, in mineral processing. In addition to the general framework for MPC, Expert Optimiser also includes tools for implementation of soft sensors and artificial intelligence applications.