Applying positive matrix factorisation to distil key information about flotation rate constants

Sandoval-Zambrano, Gerson, Whiten, William J., Gonzalo Montes-Atenas and Paatero, Pentti (2011). Applying positive matrix factorisation to distil key information about flotation rate constants. In: Willy Kracht, Romke Kuyvenhoven, Suzanne Lynch-Watson and Gonzalo Montes-Atenas, Proceedings of the 8th International Mineral Processing Seminar. Procemin 2011 - 8th International Mineral Processing Seminar, Santiago, Chile, (117-124). 30 November - 2 December 2011.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
UQ265761_fulltext.pdf HERDC full text - not publicly available Click to show the corresponding preview/stream application/pdf 1.88MB 2
Author Sandoval-Zambrano, Gerson
Whiten, William J.
Gonzalo Montes-Atenas
Paatero, Pentti
Title of paper Applying positive matrix factorisation to distil key information about flotation rate constants
Conference name Procemin 2011 - 8th International Mineral Processing Seminar
Conference location Santiago, Chile
Conference dates 30 November - 2 December 2011
Proceedings title Proceedings of the 8th International Mineral Processing Seminar
Place of Publication Santiago, Chile
Publisher Gecamin
Publication Year 2011
Sub-type Fully published paper
ISBN 9789568504625
Editor Willy Kracht
Romke Kuyvenhoven
Suzanne Lynch-Watson
Gonzalo Montes-Atenas
Start page 117
End page 124
Total pages 8
Collection year 2012
Language eng
Abstract/Summary Flotation kinetics plays an essential role in developing a comprehensive understanding of the flotation theory providing pathways for describing flotation process through mathematical modelling. Classically, the description of flotation kinetics is characterised using a coefficient,named specific flotation rate constant (k). Simultaneously, it has been acknowledged that among the number of factors affecting flotation performance, the particle properties of the ore being processed are relevant. The physical-property-based modelling approach (PPBM) Links discrete specific rate constants to physical properties of the minera l particles, such as particle size and mineral Liberation. As a result of applying the PPBM a full set of specific rate constants is obtained. In order to establish which specific rate constants are needed for prediction, it is necessary to know their uncertainties and, then, to explore, if appropriate, the existence of redundancy in the data. This work aims to apply Positive Matrix Factorisation (PMF) to condense the information of flotation rate constant values obtained from physical-property-based size- by-liberation distribution for galena in the feed stream to a 40-L continuous flotation cell. PMF uses the uncertainties associated with the rate constant values to calculate the weights of residuals. The application of PMF with a rank 1 model fitted the matrix of rate constants so that residuals of the fit were comparable to assumed data uncertainties. The two vectors followed clear trends as functions of particle size and galena liberation that may constitute heuristic for predicting flotation performance. This will be confirmed with other case studies. A simple heuristic model of the rate constant matrix might consist of weighted averages of the kinetic rate constants by using the mass proportions associated. Comparison of such a heuristic model with PMF indicates that the two models are not identical, i.e., PMF distils different information to such a heuristic model. Hence, further analysis is required.
Q-Index Code E1
Q-Index Status Confirmed Code
Institutional Status UQ

 
Versions
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Access Statistics: 116 Abstract Views, 6 File Downloads  -  Detailed Statistics
Created: Mon, 23 Jan 2012, 11:18:55 EST by Karen Holtham on behalf of Julius Kruttschnitt Mineral Research Centre