An Automated Machine learning (AutoML) approach to regression models in minerals processing with case studies of developing industrial comminution and flotation models

Koh, Edwin J.Y.; Amini, E.; Gaur, Shruti; Maquieira, Miguel Becerra; Heck, Christian Jara; McLachlan, Geoffrey J.; Beaton, Nick

doi:10.1016/j.mineng.2022.107886

Cited by 4 publications

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A similar study to the one proposed in this manuscript was performed at Los Bronces, but they maximized the throughput for a given P80 using the method AutoML. They simulated the needed throughput and accurately compared it to the actual throughput [7].…”

Section: Introductionmentioning

confidence: 99%

Machine-Learning Analysis of the Canadian Royalties Grinding Circuit

Di Feo,

Khodaie,

Girard

et al. 2024

Minerals

View full text Add to dashboard Cite

This work aimed to understand the relationships between grinding variables and the P80 (80% passing size) of a grinding circuit (feed to flotation). Canadian Royalties want to obtain and reduce variations in the P80, which is currently 65 micrometres. Thus, principal component analysis (PCA), part of machine learning, was utilized to better understand the factors that significantly influence the P80. PCA is meant to be used as a guideline for plant metallurgists to determine how the grinding circuit factors influence P80; thus, the variables can be manipulated to lower P80 fluctuations. PCA revealed that the head grade of the ore (pentlandite (Pn), chalcopyrite (Cp), pyrrhotite (Po) and non-sulphide gangue (NSG)) and the primary ball mill power were weakly correlated with P80. However, the ore level in the silo, flowrate to cyclones, cyclone pressure, percent solids and the feed tonnage rate to the primary ball mill were strongly correlated with P80. This information can be used to develop a strategy to control the P80 of the grinding circuit and provide a more consistent grind size to the flotation circuit, which can positively impact metallurgical performance.

show abstract

Section: Introductionmentioning

confidence: 99%