A kernel‐based approach for fault diagnosis in batch processes. (8th May 2014)

• Record Type: Journal Article
• Authors: Vitale, R.; de Noord, O. E.; Ferrer, A.; Åberg, Magnus
• Is Part Of: Journal of chemometrics. Volume 28:Number 8(2014:Aug.)
• Page Start: 697
• Page End: 707
• Publication Date: 2014-05-08
• Subjects: Chemistry -- Mathematics -- Periodicals; Chemistry -- Statistical methods -- Periodicals; 542.85
• DOI: 10.1002/cem.2629
• Languages: English
• ISSNs: 0886-9383

Abstract

<abstract abstract-type="main"> <title> <x xml:space="preserve">Abstract</x> </title> <p>This article explores the potential of kernel‐based techniques for discriminating on‐specification and off‐specification batch runs, combining kernel‐partial least squares discriminant analysis and three common approaches to analyze batch data by means of bilinear models: landmark features extraction, batchwise unfolding, and variablewise unfolding. Gower's idea of pseudo‐sample projection is exploited to recover the contribution of the initial variables to the final model and visualize those having the highest discriminant power. The results show that the proposed approach provides an efficient fault discrimination and enables a correct identification of the discriminant variables in the considered case studies. Copyright © 2014 John Wiley &amp; Sons, Ltd.</p> </abstract>