3750 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
transform), EMD (Empirical Mode Decomposition) and
VMD (Variational Mode Decomposition). While PSDE-
SVM was the best option for classifying different feed size
distributions, LDA was more compatible with most feature
extraction techniques for predicting feed size classes.
In contrast to the batch process previously discussed,
the study was extended to a semi-batch or locked-cycle
experiment, which is more representative of the continuous
operation used for industrial AG/SAG mill applications.
Figure 5. An illustration of feed ore size reduction and acoustics in a laboratory ball mill
Figure 6. An illustration of feed ore size reduction and acoustics in a laboratory AG/SAG mill
transform), EMD (Empirical Mode Decomposition) and
VMD (Variational Mode Decomposition). While PSDE-
SVM was the best option for classifying different feed size
distributions, LDA was more compatible with most feature
extraction techniques for predicting feed size classes.
In contrast to the batch process previously discussed,
the study was extended to a semi-batch or locked-cycle
experiment, which is more representative of the continuous
operation used for industrial AG/SAG mill applications.
Figure 5. An illustration of feed ore size reduction and acoustics in a laboratory ball mill
Figure 6. An illustration of feed ore size reduction and acoustics in a laboratory AG/SAG mill