Monitoring of an industrial liquid-liquid extraction system with kernel-based methods

Simple item page
dc.contributor.authorJemwa G.T.
dc.contributor.authorAldrich C.
dc.date.accessioned2011-05-15T16:01:40Z
dc.date.available2011-05-15T16:01:40Z
dc.date.issued2005
dc.description.abstractThe behaviour of liquid-liquid extraction systems can be complex and as a result linear methods of process condition monitoring such as principal component analysis or partial least squares may not be able to detect and identify process faults when they occur. In contrast, kernel-based methods represent a general framework that can be used where linear methods do not perform satisfactorily. This is demonstrated in a case study on an industrial liquid-liquid extraction system, where a linear discriminant analysis problem is recast as a support vector machine learning problem. The support vector machine is subsequently used to extract features from the plant data that can be used for considerably more accurate fault detection than is possible with its linear equivalent or with other nonlinear methods. Fault identification can be accomplished from an analysis of the residuals of models using the features to reconstruct the original plant data. © 2005 Elsevier B.V. All rights reserved.
dc.description.versionArticle
dc.identifier.citationHydrometallurgy
dc.identifier.citation78
dc.identifier.citation1-2 SPEC. ISS.
dc.identifier.issn0304386X
dc.identifier.other10.1016/j.hydromet.2005.03.003
dc.identifier.urihttp://hdl.handle.net/10019.1/12095
dc.subjectCorrelation methods
dc.subjectLeast squares approximations
dc.subjectMathematical models
dc.subjectMonitoring
dc.subjectNeural networks
dc.subjectNonlinear systems
dc.subjectPrincipal component analysis
dc.subjectRheology
dc.subjectVectors
dc.subjectFault detection
dc.subjectFault identification
dc.subjectKernel-based methods
dc.subjectLiquid-liquid extraction
dc.subjectProcess monitoring
dc.subjectSupport vector machines
dc.subjectSolvent extraction
dc.titleMonitoring of an industrial liquid-liquid extraction system with kernel-based methods
dc.typeArticle
Files
Collections
Stellenbosch University - Scopus Publications