4th WCSET-2015 at Japan
Civil Engineering:
Title:
Integrating artificial neural networks and geostatistics
for optimum 3D geological block modelling in mineral
reserve estimation – a case study
Authors:
Abu Bakarr Jalloh, Kyuro Sasaki
Abstract: In
this research, a procedure integrating Artificial Neural
Networks and Geostatistics for optimum mineral reserve
evaluation is presented. The method called ANNMG
(Artificial Neural Network Model integrated with
Geostatiscs) adopted the following approach: firstly,
the Artificial Neural Net was trained, tested and
validated using assay values obtained from exploratory
drillholes from the studied area. The validated model
was then used to generalize mineral grades at known
sampled and unknown locations inside the drilling region
respectively. Lastly, the reproduced and generalized
assay values were combined and fed to geostatistics for
developing a geological 3D block model. Regression
analysis reveals that the reproduced and predicted
grades by the trained neural network were in close
proximity to the actual samples. The generalized grades
from the ANNMG show that this procedure could be used to
complement exploration activities thereby reducing
drilling requirement.
Keywords: Artificial Neural
Network Model with Geostatistics (ANNMG); 3D geological
block modelling; and Mine Design; Kriging
Pages:
201-205