In recent years, the mining industry has been faced with numerous challenges across Europe and worldwide. Among these is the need to process ore with successively lower grades due to the continuous depletion of high-grade deposits. This increases the consumption of energy and water and, thus, the operational costs at a mine site. Various approaches to solve this issue have been evaluated, but so far none of these could be validated as a satisfactory solution. The implementation of multimodal sorting techniques represents a promising approach by achieving a pre-concentration of valuable minerals already at an early stage in the metallurgical process.
In this project we propose to develop a fusion technology including laser-induced breakdown spectroscopy (LIBS) and multi energy X-ray transmission (ME-XRT), which will be able to classify crushed mineral particles on a conveyor belt with the aid of deep learning technology. The combination of LIBS and ME-XRT is promising, as these sensors complement each other with regards to their analytical capabilities: LIBS can provide an elemental analysis of the sample surface, while ME-XRT produces volumetric data with lower accuracy. The technological fusion of both sensors will allow for the extrapolation of accurate surface data to the entire volume of the sample and therefore create representative data for the entire ore. In addition, the implementation of neural network technology will enable allow for automatic self-adjustments to varying ore types and geological parameters.
The developed sensor fusion technology will enable constant and accurate monitoring of the mineralogy of the mined rock volume and will allow for on-line and in-situ measurement of geological, mineralogical, rock-mechanical and metallurgical properties of the ore. The development of an on-line feed of these data into 3D geological models of the ore bodies is envisaged, the accuracy and objectivity of which are crucial for successful mine planning.