Neural nets for the simulation of mineral processing operations: Part II. Applications

Simple item page
dc.contributor.authorVan der Walt T.J.
dc.contributor.authorVan Deventer J.S.J.
dc.contributor.authorBarnard E.
dc.date.accessioned2011-05-15T15:54:00Z
dc.date.available2011-05-15T15:54:00Z
dc.date.issued1993
dc.description.abstractThis paper shows that neural nets exhibit exceptional promise as modelling tool and can be applied and developed further for various applications in the metallurgical processing industry. It is described how a sigmoidal backpropagation neural network (SBNN) model for the classification efficiency of a hydrocyclone classifier can be developed on the basis of sufficient data. However, data are expensive and difficult to obtain for many systems in the processing industry. As difficulties are encountered if a nonparametric model is constructed on the basis of sparse data, a new neural network modelling technique is described to obviate this problem. The hybrid subspace method has been developed to isolate the dimensions of less-significant variables and to identify some mathematical relations, so that the ill-defined dimensionality is reduced and the population density of data is increased accordingly. It has been found that the performance of a hybrid subspace model for the kinetics of a typical processing operation is superior to that of an SBNN model for the entire predictor variable space. © 1993.
dc.description.versionArticle
dc.identifier.citationMinerals Engineering
dc.identifier.citation6
dc.identifier.citation11
dc.identifier.issn8926875
dc.identifier.urihttp://hdl.handle.net/10019.1/8937
dc.titleNeural nets for the simulation of mineral processing operations: Part II. Applications
dc.typeArticle
Files
Collections
Stellenbosch University - Scopus Publications