A multi-input-single-output smith predictor for feeders control in SAG grinding plants

The accurate control of feeders in semiautogenous grinding(SAG) mills plays an important role for obtaining good operational results. This brief presents the design of a Smith predictor for multi-input-single-output (MISO) systems, representing a set of feeders in a SAG plant. The design considers the special characteristics of the application, as well as issues like performance and robustness of the proposed controller. Experiments carried out at an industrial plant, under several operational conditions, demonstrate the effectiveness of the proposed strategy compared to the one based on a single proportional-integral derivative (PID) controller.