0
Research Papers: Nuclear Power

Detection and Identification of Faults in NPP Instruments Using Kernel Principal Component Analysis

[+] Author and Article Information
Jianping Ma

Department of Electrical and Computer Engineering,  University of Western Ontario, London, Ontario, N6A 5B9, Canadajma64@uwo.ca

Jin Jiang

Department of Electrical and Computer Engineering,  University of Western Ontario, London, Ontario, N6A 5B9, Canada; Institute of Nuclear Science and Technology,  Xi’an Jiaotong University, Xi’an, Shaanxi, 710049, Chinajjiang@uwo.ca

J. Eng. Gas Turbines Power 134(3), 032901 (Dec 29, 2011) (6 pages) doi:10.1115/1.4004596 History: Received July 23, 2010; Revised July 26, 2010; Published December 29, 2011; Online December 29, 2011

In this paper, kernel principal component analysis (KPCA) is studied for fault detection and identification of the instruments in nuclear power plants. A KPCA model for fault isolation and identification is proposed by using the average sensor reconstruction errors. Based on this model, faults in multiple sensors can be isolated and identified simultaneously. Performance of the KPCA-based method is demonstrated with real NPP measurements.

FIGURES IN THIS ARTICLE
<>
Copyright © 2012 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Principle of kernel PCA

Grahic Jump Location
Figure 2

Fault detection results

Grahic Jump Location
Figure 3

Average reconstruction errors (absolute value) for data sets 3-6

Grahic Jump Location
Figure 4

Average reconstruction errors (percentage) for data sets 3-6

Grahic Jump Location
Figure 5

Mean reconstruction errors for data sets 6

Grahic Jump Location
Figure 6

Fault indices for selected sensors

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In