Advanced Control Optimization in Copper Concentrators for Better Mineral Throughput and Size Classification

Additonal authors: Nazari, S.. Book title: Proceedings of the 58th Conference of Metallurgists Hosting Copper 2019. Chapter: . Chapter title:

Proceedings, Vol. Proceedings of the 58th Conference of Metallurgists Hosting Copper 2019, 2019

Carricajo T.

This paper discusses a unique dynamic solution for optimizing comminution area in real time. The proposed technology is a combination of model predictive controller, optimization methods and real-time digital twin for mineral processing. For grinding area, the ultimate goal of an effective control strategy is to maximize the mineral processing, reduce energy consumption and stabilized size classification. To have a continuous and effective comminution operation, a reliable control system that minimizes the need for human intervention is necessary. The implemented advanced control strategy for comminution area consists of two layers in hierarchical structure. The lower layer objective is to stabilize the process stabilization using model based predictive controller (MPC) technique. The unique MPC used in this layer is called BrainWave™ chosen for its robustness, convergence and delay handling; furthermore, the developed platform offers stable and repeatable solution and has short implementation time. The upper layer is a supervisory control, which sends the preferred process operation points to the lower layer to drive the process towards optimal performance. The goal of this layer is to optimize the process targets, as well as making decision. To increase process performance, automatic ore characteristic adaptations are implemented in this layer. A digital twin of comminution area is built in the second layer that runs amongst the process in parallel. By communicating, the process data in real time, a virtual densitometer is built that is being used in our advance process control strategy. Our control strategy has been developed, tested and implemented in various areas of a copper concentration plant. The targets for implementing advance process control are usually increased throughput, decrease energy consumption and variability reduction in particle size distribution. Overall, the results showed in different mine sites are increase throughput between 2 and 5%, specific energy consumption reduction about 15% and process variability reduction up to 60%. In this paper we discuss the results of the recent implementation of this technology on Minera Los Pelambres, Minera Collahuasi and CAP Minería. Their concentrator plants have installed control systems with the most updates optimization features, for SAG mill, ball mills and thickeners.
Mots Clés: Copper 2019, COM2019