Supervised learning algorithms for estimation of liners wear in SAG Mills

Abstract

The measurement of wear in the liners of SAG mills is performed with the machine stopped and entering inside in a short available stopping time. Thus, the height estimation of liners will save time and costs when liners are changed. This paper proposes an estimator of liners height in SAG mills from three-stage process data using Machine Learning algorithms for supervised learning. Two models of estimation were proposed and trained based on the information acquired and processed by the SCADA system of the process.