Constrained real-time optimization of a grinding circuit using steady-state linear programming supervisory control

Hardware-in-the-loop simulation platform for supervisory control of mineral grinding process

“…Until now, there have been some attempts on solving this supervisory control problem. These begin with some model-based control and optimization methods, such as real-time optimization (RTO) [25], model predictive control (MPC) [2,[26][27][28] and adaptive decoupled control [29,30]. But, these methods are hard to be applied in practical MGPs, as accurate modeling is difficult to achieve or the established models do not accurately describe the actual dynamic processes.…”

Systems optimization model for energy management of a parallel HPGR crushing process

“…Since both crushing and grinding machines behave generally as constant loads, their torque characteristics (torque versus speed) is almost constant for a given loading level. 2 Knowing that the mechanical power is proportional to the rotational speed, the mechanical power of the HPGR can be therefore approximated to a linear function of the speed with the proportional constant being the torque demand. This is proven by Eq.…”

“…In Ref. [2], a linear programming supervisory control is employed to maximize the grinding circuit throughput, while in Ref. [3], the grinding circuit throughput is maximized using an expert system based on fuzzy logic where an increase of 10% in feed tonnage is achieved.…”

“…[4,5], the cost function to be maximized is taken as the grinding circuit profit instead of the grinding circuit throughput as in Refs. [1,2,3]. The Hooke and Jeeves search routine algorithm is used in Ref.…”

“…[7,8,10,15] are those where open loop systems optimization is applied, while in Refs. [1,2,3,4,5,6,9,11,12,13,14,16], a closed-loop systems 3 optimization is applied.…”