XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 3913
soft on SAG mill performance. These components were
quantified based on the novel multi-component modelling
approach detailed in section 5 of this paper. Figure 8 shows
the configured flowsheet with three feeders assigned to each
component.
The main objective of these simulations is to estimate
the impact of competence variation on SAG mill perfor-
mance. The breakage model parameters (A and b) estimated
for hard, medium and soft components given in Table 4
were used as inputs to the SAG mill model. A change in
the SAG mill performance will have an impact on the sub-
sequent performance of equipment such as the ball mill
and hydrocyclone. However, this study did not take those
impacts into consideration. It should be noted that the con-
siderations in this study are as follows:
The simulated SAG mill load variation should be con-
strained within a narrow range of ±0.3%. Bailey et al. (2009)
highlighted that a change in ore competence will associate
with changes in the AG/SAG mill filling/charge. It should
be noted that the AG/SAG mill model does not accom-
modate variation of the breakage rates with the mill load.
It should however be noted that the SAG mill model out-
comes are most reliable at filling close to the measured load.
The feed size distributions for different components
were assumed to remain the same, whereas in reality the
feed size distribution of different components is depen-
dent on ore characteristics (structure and competence)
and blast design parameters It should be acknowledged
that components with different competence in feed having
distinguishable size distributions, in particular when these
materials are blended. However, it should be emphasised
that in this paper we focus on quantifying ore competence
variability and estimate proportion of components within
one ore type that differs from the case of blending ‘different
ore types’. Having an estimate of how different components
of ‘an ore type’ are distributed within the feed, should allow
to more realistically quantify performance variation from
the feed (its competence and size distribution)—that may
increase predicted variations. However, this likely phenom-
enon is not addressed in this paper and requires further
investigation.
The actual SAG mill ball charge (Jb) was 12.1%.
However, to further investigate the impact of this vari-
able on ore breakdown in a multi-component space, addi-
tional simulations were conducted with 5%, 10% and 15%
ball charges.
This paper explores 13 scenarios through three sets of
defined simulations, which are outlined as follows:
Combinatorial multi-component simulation: This
includes a range of simulations with changing proportion
of hard, medium and soft components within Sample A
ore type with breakage parameters (A, b) given in Table 3.
Different proportion of competences are denoted as a Hard
:Medium :Soft format that adds up to 100, e.g., 25:50:25.
In this essence, nine scenarios of 90:5:5, 80:10:10,
60:20:20, 5:90:5, 10:80:10, 20:60:20, 5:5:90, 10:10:80,
Figure 8. The SAG milling configuration used for simulation scenarios
soft on SAG mill performance. These components were
quantified based on the novel multi-component modelling
approach detailed in section 5 of this paper. Figure 8 shows
the configured flowsheet with three feeders assigned to each
component.
The main objective of these simulations is to estimate
the impact of competence variation on SAG mill perfor-
mance. The breakage model parameters (A and b) estimated
for hard, medium and soft components given in Table 4
were used as inputs to the SAG mill model. A change in
the SAG mill performance will have an impact on the sub-
sequent performance of equipment such as the ball mill
and hydrocyclone. However, this study did not take those
impacts into consideration. It should be noted that the con-
siderations in this study are as follows:
The simulated SAG mill load variation should be con-
strained within a narrow range of ±0.3%. Bailey et al. (2009)
highlighted that a change in ore competence will associate
with changes in the AG/SAG mill filling/charge. It should
be noted that the AG/SAG mill model does not accom-
modate variation of the breakage rates with the mill load.
It should however be noted that the SAG mill model out-
comes are most reliable at filling close to the measured load.
The feed size distributions for different components
were assumed to remain the same, whereas in reality the
feed size distribution of different components is depen-
dent on ore characteristics (structure and competence)
and blast design parameters It should be acknowledged
that components with different competence in feed having
distinguishable size distributions, in particular when these
materials are blended. However, it should be emphasised
that in this paper we focus on quantifying ore competence
variability and estimate proportion of components within
one ore type that differs from the case of blending ‘different
ore types’. Having an estimate of how different components
of ‘an ore type’ are distributed within the feed, should allow
to more realistically quantify performance variation from
the feed (its competence and size distribution)—that may
increase predicted variations. However, this likely phenom-
enon is not addressed in this paper and requires further
investigation.
The actual SAG mill ball charge (Jb) was 12.1%.
However, to further investigate the impact of this vari-
able on ore breakdown in a multi-component space, addi-
tional simulations were conducted with 5%, 10% and 15%
ball charges.
This paper explores 13 scenarios through three sets of
defined simulations, which are outlined as follows:
Combinatorial multi-component simulation: This
includes a range of simulations with changing proportion
of hard, medium and soft components within Sample A
ore type with breakage parameters (A, b) given in Table 3.
Different proportion of competences are denoted as a Hard
:Medium :Soft format that adds up to 100, e.g., 25:50:25.
In this essence, nine scenarios of 90:5:5, 80:10:10,
60:20:20, 5:90:5, 10:80:10, 20:60:20, 5:5:90, 10:10:80,
Figure 8. The SAG milling configuration used for simulation scenarios