Multivariate Mapping of Geometallurgical Variables with Complex Cross-Correlation Characteristics

Project: Research project

Grant Program

Faculty Development Competitive Research Grant Program 2018-2020

Project Description

The rational of this 3-years project is to establish a creative workflow enabling multivariate geometallurgical mapping for enhanced characterization of mineral resources for mining industry in Kazakhstan. The current methodology for multivariate geostatistical modeling of the geometallurgical variables often lack realism from the complex cross-correlation characteristics and often leads to impractical results. We propose to develop and expand a new framework entitled “Enhanced Projection Pursuit Multivariate Transform (EPPMT)” for multivariate geometallurgical mapping such as 3D modeling of comminution indices (e.g., bond rod mill work index, resistance to abrasion and breakage index, drop-weight index and some other rock strength properties), that quite often show complex cross-correlation dependencies (heteroscedasticity, Geological constraint, Non-linearity).

Key findings

Papers (Journals and Conferences)
Short titleMMGVCCS
StatusActive
Effective start/end date3/20/1812/31/20

Fingerprint

transform
simulation
modeling
kriging
methodology
nonlinearity
comminution
mining industry
autocorrelation
industrial mineral
rock
variogram
mineral
mineral resource
metal
zinc
method
copper
engineering
rhenium