Abstract
The challenges in mineral processing and productivity are well known. Mostly caused by grade decline and complex mineralogy, miners might experience difficulty with process optimization and stability in mineral processing plants.
Additionally, there are limits of basic control strategies in processing plants, given the non-linearity of the process, recirculating loads and multiple interactions. Basic control strategies rarely achieve optimal control in terms of quality, nor does it operate in the most cost-effective manner.
This presentation will explain the basics of analytics. It will cover how advanced technologies such as machine learning, model predictive control, virtual analyzers and anomaly detection can be used for process optimization, reduced variability, reduced specific energy consumption, and to identify abnormalities in mineral processing operations. More efficient mineral processing not only leads to increased productivity, but also allows plants to operate at a higher capacity and lean towards capital avoidance.