Open Collections

Wong, Benjamin

University of British Columbia

To fuel the transition to a green economy, the demand for critical elements, such as copper, have seen a sharp increase. Mining techniques such as large-scale block caving have been utilized to meet this mineral demand. To improve block caving efficiency, technologies such as X-ray fluorescence (XRF) sensors aim to improve efficiency through grade monitoring. This paper evaluated the case for sensor-based grade monitoring at the Deep Mill Level Zone (DMLZ) mine in the Grasberg Mineral Complex. Three separate studies were done. Study 1 evaluated variability/heterogeneity of the DMLZ ore body by comparing the PCBC model and drawpoint assay data provided by PTFI. The study found that both the PCBC projections and the drawpoint assay data disagreed with respect to mineral grades and variability, with drawpoint assay data showing considerably more variance. This lack of agreement between PCBC and the assay data suggested that improved grade monitoring would allow for an additional data point when the other two sources of information do not agree. Additionally, certain areas of the mine showed higher levels of overall variability in both datasets, suggesting that these areas could benefit from prioritised grade monitoring. Study 2 quantified the confidence that sensor-based grade monitoring equipment would provide compared to current methods. Sampling theory recommends using sample sizes measured in tonnes. This scale is possible for sensor-based grade monitoring equipment, but unrealistic with current grade monitoring techniques. A limitation of this study was that the analytical error variation between current and proposed techniques was unable to be analysed. The final study attempted to model gold grades with elements detectable by XRF sensors. Simple linear regression models were compared to complex models utilising machine learning. After comparing the performance of both models, the conclusion drawn was that the elements available to study did not improve gold modelling by any appreciable quantity, though there are opportunities for additional study. Overall, there is a strong case made for trials into sensor-based grade monitoring equipment. However, additional study is recommended to quantify the benefit of using such equipment, as sorting waste from ore at the mine is currently not considered possible.

Text

2025-01-09

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/90079

Mining Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/