888 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
velocity (b) to keep the controlled variables close to the ref-
erence setpoint. Changes in both variables stand out for
occurring smoothly over the analyzed period, thus ensuring
stable operation without risking the mechanical structure
of the equipment. Analysis of Figure 5 in conjunction with
Figure 4 also allows us to conclude that limitations in main-
taining controlled variables close to the reference setpoints
after 140 minutes of operation (Figure 4) are due to the fact
that the machine reached the lower pressure limit (40 bar)
when there was a significant increase in the Blaine specific
surface area of the feed, limiting the controller’s efficiency
for this time period.
In analogy to case study #1, Figure 6 presents the evo-
lution of the product BSA (a) and throughput (b) for case
study #2, where variations in the feed BSA also followed
the dynamics shown in Figure 2b, now with an increase in
the required product quality (reference setpoint equal to
1,950 cm2/g) and throughput (reference setpoint equal to
700 t/h). Results in Figure 6a highlight the limitations of
the NMPC to ensure a product close to 1,950 cm2/g, with
Figure 4. Response of the product BSA (a) and throughput (b) in the 180 min time window for case study #1. Changes in the
HPGR feed BSA (process disturbance) followed predictions made by Eq. (21) and presented in Figure 2b
Figure 5. Response of the operating pressure and gap (a) and roll peripheral velocity (b) in 180 min time window for case
study #1. Changes in the HPGR feed BSA (process disturbance) followed predictions made by Eq. (21) and presented in
Figure 2b
velocity (b) to keep the controlled variables close to the ref-
erence setpoint. Changes in both variables stand out for
occurring smoothly over the analyzed period, thus ensuring
stable operation without risking the mechanical structure
of the equipment. Analysis of Figure 5 in conjunction with
Figure 4 also allows us to conclude that limitations in main-
taining controlled variables close to the reference setpoints
after 140 minutes of operation (Figure 4) are due to the fact
that the machine reached the lower pressure limit (40 bar)
when there was a significant increase in the Blaine specific
surface area of the feed, limiting the controller’s efficiency
for this time period.
In analogy to case study #1, Figure 6 presents the evo-
lution of the product BSA (a) and throughput (b) for case
study #2, where variations in the feed BSA also followed
the dynamics shown in Figure 2b, now with an increase in
the required product quality (reference setpoint equal to
1,950 cm2/g) and throughput (reference setpoint equal to
700 t/h). Results in Figure 6a highlight the limitations of
the NMPC to ensure a product close to 1,950 cm2/g, with
Figure 4. Response of the product BSA (a) and throughput (b) in the 180 min time window for case study #1. Changes in the
HPGR feed BSA (process disturbance) followed predictions made by Eq. (21) and presented in Figure 2b
Figure 5. Response of the operating pressure and gap (a) and roll peripheral velocity (b) in 180 min time window for case
study #1. Changes in the HPGR feed BSA (process disturbance) followed predictions made by Eq. (21) and presented in
Figure 2b