Abstract
The mining industry is facing increasing pressure to improve sustainability through reducing operations' GHG emissions, water and power consumption, and chemicals’ usage. This can be achieved through mine-to-mill optimization with MineSense Technologies ShovelSense© and BeltSense© systems. MineSense’s sensor-based technology uses X-ray fluorescence (XRF) to analyze ore in real-time as it is mined. ShovelSense can provide the elemental composition of the ore at a bucket-by-bucket resolution, which can be used to determine the most efficient methods for extracting ore. Additionally, by combining this data with real-time measurements in the grinding circuit using BeltSense, it is possible to improve recovery and reduce water, energy intensity, and reagent consumption.
Here we present the results of a year of data collection at an open-pit copper mine equipped with ShovelSense on the entire mining fleet and BeltSense on two belts within the grinding circuit. Mine-to-mill optimization using this real-time XRF sensor-based ore sorting technology improved the overall grade of mined ore by replacing waste rock with high-grade ore and reduced dilution leading to a Cu unit increase of 7% and an improved ore recovery of 7.7%. Additionally, MineSense has developed machine learning algorithms to further characterize the ore mineral composition and grindability. These predictions can be used to optimize the flotation process and SAG Mill further.