Fault Detection for Process Monitoring Using Improved Kernel Principal Component Analysis

In order to detect abnormal events of chemical processes, a new fault detection method based on kernel principal component analysis (KPCA) is described. Firstly, it removes the noise from data set using wavelet packet transform (WPT). Secondly, a feature vector selector schemes (FVS) based on a geom...

Full description

Saved in:
Bibliographic Details
Published in2009 International Conference on Artificial Intelligence and Computational Intelligence Vol. 2; pp. 334 - 338
Main Authors Jie Xu, Shousong Hu, Zhongyu Shen
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.11.2009
Subjects
Automation
Chemical processes
Educational institutions
Fault detection
Feature vector selection
Kernel
Kernel principal component analysis
Monitoring
Principal component analysis
Signal analysis
wavelet analysis
Wavelet packets
Wavelet transforms
Online AccessGet full text

Cover

More Information
Summary:In order to detect abnormal events of chemical processes, a new fault detection method based on kernel principal component analysis (KPCA) is described. Firstly, it removes the noise from data set using wavelet packet transform (WPT). Secondly, a feature vector selector schemes (FVS) based on a geometrical consideration is given to reduce the computation complexity of KPCA when the number of the samples becomes large. Then, the denoised data is disposed using KPCA and and SPE are constructed in the feature space. KPCA was applied to fault detection. To demonstrate the performance, the proposed method is applied to the Tennessee Eastman process. The simulation results show that the proposed method effectively and quickly detect various types of faults.
ISBN:1424438357
9781424438358
DOI:10.1109/AICI.2009.43