XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1043
copper mine in Central Chile, incorporating various opera-
tions such as blasting, crushing, grinding, and flotation.
Industry-standard phenomenological models were applied
to blasting and comminution process units. Additionally,
for the flotation operation, a dedicated machine learning
predictive model was developed to forecast elemental recov-
ery and grade based on changes in ore properties.
The developed flow sheet underwent validation against
historical data, demonstrating the reliability of the out-
comes. It opens up opportunities for short to medium-
term plant optimization efforts without disrupting ongoing
operations, providing effective assessments of crises during
short-term challenges, and facilitating short-term optimiza-
tion initiatives. Moreover, the flow sheet supports medium-
term optimization analyses, including the optimization of
blending stockpile and fresh ore for enhanced performance.
Specifically, the flowsheet was employed for the technical
assessment of an additional grinding line in the comminu-
tion circuit and its impact on metal production. The simu-
lation indicates a noteworthy 25.3% increase in throughput
with the incorporation of an additional grinding line into
the existing circuit. This improvement, however, comes at
the expense of a marginal increase in specific energy and
daily water consumption. Availability process water in dry
season was considered alongside other constraints in simu-
lations and production forecasting.
REFERENCE
Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean,
J., &Yu, Y., 2016. Tensorflow: A System for Large-
Scale Machine Learning. In Proceedings of the 12th
USENIX Symposium on Operating Systems Design
and Implementation. Savannah, USA.
Amini, E., Becerra, M., Bachmann, T., Beaton, N., &
Shapland, G., 2020. Development and Reconciliation
of a Mine Operation Value Chain Flow sheet in
IES to Enable Grade Engineering and Process Mass
Simulations for Scale-Up and Strategic Planning
Analysis. In SME Annual Conference and Expo,
Arizona, February 23–26, 2020.
Amini, E., &Beaton, N., 2020. Development of mine
operation value chain flow sheets fine-tuned and con-
straint based on the process information (PI) data to
evaluate proposed operation strategies and ore pre-
treatment technologies. In Preconcentration Digital
Conference, November 17–18, 2020.
Andersen, J.S., &Napier-Munn, T.J., 1988. Power pre-
diction of cone crushers. In Proc. 3rd Mill Operators’
Conference, AusIMM.
Carrasco, C., Gahona, M., La Rosa, D., Shapland,
G., Beaton, N., &Amini, E., 2019. Augmenting
Traditional Mine to Mill with Cloud-Based Simulation
and Reconciliation Capabilities. In SAG Conference,
Vancouver, 17–19 January 2019.
Chollet, F., 2015. Keras. Retrieved from GitHub reposi-
tory: https://github.com/fchollet/keras.
Cunningham, C.V.B., 1983. The Kuz–Ram model for pre-
diction of fragmentation from blasting. Proceedings of
1st International Symposium on Rock Fragmentation
by Blasting, Lulea, pp 439–454.
Kanchibotla, SS., Valery, W., &Morell, S., 1999. Modelling
fines in blast fragmentation and its impact on crushing
and grinding. In Proceedings of the Explo 1999 con-
ference. Carlton, Victoria: AusIMM p. 137–44.
Keeney, L., Beaton, N., Scott, M., La Rosa, D., Amini, E.,
Rutter, J., &Faramarz, F., 2020. The integration of
mine to mill and grade engineering at Anglo American’s
Los Bronces Mine using IES. In Preconcentration
Digital Conference, November 17–18, 2020.
Koh, E.J., Amini, E., Watkins, R., Mayhungu, E., &
Bergmann, C., 2023. Maximising Value at a Platinum
Group Metals Concentrator by Development of an
Integrating Comminution and Flotation Flow sheet.
In comminution 2023, Cape Town, South Africa.
Koh, E.J., Amini, E., McLachla, C.J., &Beaton, N.,
2021. Utilising a deep neural network as a surrogate
model to approximate phenomenological models of a
comminution circuit for faster simulations. Minerals
Engineering, 170, 107026. ISSN 0892-6875.
Morrell, S., Valery, W., Banini, G., &Latchireddi, S.,
2001. Development in AG/SAG Mill Modelling. SAG
Conference, Vancouver, pp. 71–84.
Table 2. Comparison plant output for adding another grinding line to the existing circuit
Number of
Grinding Lines
Throughput,
t/h
Grind Size,
µm Total Power, kW
Total Specific
Energy, kWh/t
Water Consumption Rate,
m3/t
1–3 6980 204 88973 12.74 1.9
1–4 8740 204 114912 13.15 1.9
copper mine in Central Chile, incorporating various opera-
tions such as blasting, crushing, grinding, and flotation.
Industry-standard phenomenological models were applied
to blasting and comminution process units. Additionally,
for the flotation operation, a dedicated machine learning
predictive model was developed to forecast elemental recov-
ery and grade based on changes in ore properties.
The developed flow sheet underwent validation against
historical data, demonstrating the reliability of the out-
comes. It opens up opportunities for short to medium-
term plant optimization efforts without disrupting ongoing
operations, providing effective assessments of crises during
short-term challenges, and facilitating short-term optimiza-
tion initiatives. Moreover, the flow sheet supports medium-
term optimization analyses, including the optimization of
blending stockpile and fresh ore for enhanced performance.
Specifically, the flowsheet was employed for the technical
assessment of an additional grinding line in the comminu-
tion circuit and its impact on metal production. The simu-
lation indicates a noteworthy 25.3% increase in throughput
with the incorporation of an additional grinding line into
the existing circuit. This improvement, however, comes at
the expense of a marginal increase in specific energy and
daily water consumption. Availability process water in dry
season was considered alongside other constraints in simu-
lations and production forecasting.
REFERENCE
Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean,
J., &Yu, Y., 2016. Tensorflow: A System for Large-
Scale Machine Learning. In Proceedings of the 12th
USENIX Symposium on Operating Systems Design
and Implementation. Savannah, USA.
Amini, E., Becerra, M., Bachmann, T., Beaton, N., &
Shapland, G., 2020. Development and Reconciliation
of a Mine Operation Value Chain Flow sheet in
IES to Enable Grade Engineering and Process Mass
Simulations for Scale-Up and Strategic Planning
Analysis. In SME Annual Conference and Expo,
Arizona, February 23–26, 2020.
Amini, E., &Beaton, N., 2020. Development of mine
operation value chain flow sheets fine-tuned and con-
straint based on the process information (PI) data to
evaluate proposed operation strategies and ore pre-
treatment technologies. In Preconcentration Digital
Conference, November 17–18, 2020.
Andersen, J.S., &Napier-Munn, T.J., 1988. Power pre-
diction of cone crushers. In Proc. 3rd Mill Operators’
Conference, AusIMM.
Carrasco, C., Gahona, M., La Rosa, D., Shapland,
G., Beaton, N., &Amini, E., 2019. Augmenting
Traditional Mine to Mill with Cloud-Based Simulation
and Reconciliation Capabilities. In SAG Conference,
Vancouver, 17–19 January 2019.
Chollet, F., 2015. Keras. Retrieved from GitHub reposi-
tory: https://github.com/fchollet/keras.
Cunningham, C.V.B., 1983. The Kuz–Ram model for pre-
diction of fragmentation from blasting. Proceedings of
1st International Symposium on Rock Fragmentation
by Blasting, Lulea, pp 439–454.
Kanchibotla, SS., Valery, W., &Morell, S., 1999. Modelling
fines in blast fragmentation and its impact on crushing
and grinding. In Proceedings of the Explo 1999 con-
ference. Carlton, Victoria: AusIMM p. 137–44.
Keeney, L., Beaton, N., Scott, M., La Rosa, D., Amini, E.,
Rutter, J., &Faramarz, F., 2020. The integration of
mine to mill and grade engineering at Anglo American’s
Los Bronces Mine using IES. In Preconcentration
Digital Conference, November 17–18, 2020.
Koh, E.J., Amini, E., Watkins, R., Mayhungu, E., &
Bergmann, C., 2023. Maximising Value at a Platinum
Group Metals Concentrator by Development of an
Integrating Comminution and Flotation Flow sheet.
In comminution 2023, Cape Town, South Africa.
Koh, E.J., Amini, E., McLachla, C.J., &Beaton, N.,
2021. Utilising a deep neural network as a surrogate
model to approximate phenomenological models of a
comminution circuit for faster simulations. Minerals
Engineering, 170, 107026. ISSN 0892-6875.
Morrell, S., Valery, W., Banini, G., &Latchireddi, S.,
2001. Development in AG/SAG Mill Modelling. SAG
Conference, Vancouver, pp. 71–84.
Table 2. Comparison plant output for adding another grinding line to the existing circuit
Number of
Grinding Lines
Throughput,
t/h
Grind Size,
µm Total Power, kW
Total Specific
Energy, kWh/t
Water Consumption Rate,
m3/t
1–3 6980 204 88973 12.74 1.9
1–4 8740 204 114912 13.15 1.9