XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1777
close to their respective targets throughout the examined
period.
In this analysis, the control performance is also quanti-
fied by the root mean squared error (RMSE) between the
PV and SP, and the RMSE formula is defined by
RMSE N
PV /^SP
=
-h2
where N represents the length of the sample over which the
RMSE is calculated. Table 1 summarizes the RMSE values
of the residual cyanide concentration and pH level in the
CIL Tank 1.
CONCLUSIONS
For mineral processing, conventional control loops and
expert control systems frequently encounter challenges in
delivering desirable performance, particularly when deal-
ing with control problems characterized by long dead time,
nonlinearity, and multivariate relationships. These inherent
complexities in process dynamics often restrict the effec-
tiveness of traditional control strategies.
Consequently, the implementation of MPCs is critical
for optimizing mineral processing operations. ANDRITZ’s
APC solutions use the BrainWave to perform process con-
trol optimization in various areas of mineral processing.
This includes crucial applications in the CIL process of
gold extraction. In the analyzed CIL circuit, the APC solu-
tion demonstrated its effectiveness by consistently main-
taining the residual cyanide concentration and pH level
close to their respective targets and constraining the RMSE
Figure 4. Residual cyanide concentration regulated by APC in October 2023
Figure 5. pH level regulated by APC in October 2023
Table 1. CIL ACE APC performance summary
RMSE APC Utilization, %
Residual cyanide
concentration
0.116 100
pH level 0.102 100
close to their respective targets throughout the examined
period.
In this analysis, the control performance is also quanti-
fied by the root mean squared error (RMSE) between the
PV and SP, and the RMSE formula is defined by
RMSE N
PV /^SP
=
-h2
where N represents the length of the sample over which the
RMSE is calculated. Table 1 summarizes the RMSE values
of the residual cyanide concentration and pH level in the
CIL Tank 1.
CONCLUSIONS
For mineral processing, conventional control loops and
expert control systems frequently encounter challenges in
delivering desirable performance, particularly when deal-
ing with control problems characterized by long dead time,
nonlinearity, and multivariate relationships. These inherent
complexities in process dynamics often restrict the effec-
tiveness of traditional control strategies.
Consequently, the implementation of MPCs is critical
for optimizing mineral processing operations. ANDRITZ’s
APC solutions use the BrainWave to perform process con-
trol optimization in various areas of mineral processing.
This includes crucial applications in the CIL process of
gold extraction. In the analyzed CIL circuit, the APC solu-
tion demonstrated its effectiveness by consistently main-
taining the residual cyanide concentration and pH level
close to their respective targets and constraining the RMSE
Figure 4. Residual cyanide concentration regulated by APC in October 2023
Figure 5. pH level regulated by APC in October 2023
Table 1. CIL ACE APC performance summary
RMSE APC Utilization, %
Residual cyanide
concentration
0.116 100
pH level 0.102 100