1566 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
SiO2
in concentrate
..2407 *
.*
*
.57 */Fe
*
eK+0.77
GLh
GL/Fe GLh
GLh
GL/Fe GLh
0 321 0
395 1/Fe
9 1
/G1hN GL
=
+J-
L
K
K+0
K
K K+
K ^PPC
^PPC
^G
^G
^Fe
P
O
O
O
O
O
O
O
(4)
Spatial Geometallurgical Model
The generation of large-scale geometallurgical 3D mod-
els for ore bodies presents a novel and pivotal challenge
for mathematical geosciences, necessitating innovative
advancements (Van den Boogart and Tolosana-Delgado,
2018). Geostatistical and other numerical techniques are
being refined and expanded to create these high-resolution
models, incorporating a comprehensive range of available
data. Key considerations involve addressing disparities in
sampling between metallurgical properties and grade assays,
accommodating measurements at varying scales, and effec-
tively managing the intricate nonlinear averaging of numer-
ous metallurgical parameters (Deutsch et al., 2016).
In this study, transfer functions were applied for spatial
geometallurgical modeling, leveraging the development of
Figure 7. Regression tree model for SiO2 content estimation in ABO concentrate