Fault Detection of Non-Linear Processes Using Kernel Independent Component Analysis

Jong-Min LEE, S. Joe QIN*, In-Beum LEE

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review

209 Citations (Scopus)

Abstract

In this paper, a new non-linear process monitoring method based on kernel independent component analysis (KICA) is developed. Its basic idea is to use KICA to extract some dominant independent components capturing non-linearity from normal operating process data and to combine them with statistical process monitoring techniques. The proposed method is applied to the fault detection in the Tennessee Eastman process and is compared with PCA, modified ICA, and KPCA. The proposed approach effectively captures the non-linear relationship in the process variables and showed superior fault detectability compared to other methods while attaining comparable false alarm rates.
Original languageEnglish
Pages (from-to)526-536
Number of pages11
JournalCanadian Journal of Chemical Engineering
Volume85
Issue number4
DOIs
Publication statusPublished - Aug 2007
Externally publishedYes

Keywords

  • Fault detection
  • Kernel independent component analysis (KICA)
  • Non-linear component analysis
  • Process monitoring
  • principal component analysis (PAC)

Fingerprint

Dive into the research topics of 'Fault Detection of Non-Linear Processes Using Kernel Independent Component Analysis'. Together they form a unique fingerprint.

Cite this