Robust nonlinear model predictive control of a run-of-mine ore milling circuit

Abstract

This brief investigates the feasibility of applying a robust nonlinear model predictive controller to a run-of-mine ore milling circuit, and the conditions under which such a controller might be worthwhile implementing. The run-of-mine ore milling circuit model used consists of nonlinear modules for the individual components of the milling circuit, allowing for arbitrary milling circuit configurations to be modeled. The model is cast into a robust nonlinear model predictive control framework, and a practically motivated simulation of the mill model being controlled by an robust nonlinear model predictive control (RNMPC) controller is presented. Issues related to implementing such a controller are investigated.