Fault Detection and Identification in NPP Instruments Using Kernel Principal Component Analysis
Keyword(s):
In this paper, kernel principal component analysis (KPCA) is studied for fault detection and identification in the instruments of nuclear power plants. We propose to use mean values of the sensor reconstruction errors of a KPCA model for fault isolation and identification. They provide useful information about the directions and magnitudes of detected faults, which are usually not available from other fault isolation techniques. The performance of the method is demonstrated by applications to real NPP measurements.
2011 ◽
Vol 134
(3)
◽
2020 ◽
pp. 014233122096024
2016 ◽
Vol 40
(4)
◽
pp. 1289-1296
◽
2014 ◽
Vol 31
(6)
◽
pp. 930-943
◽
2013 ◽
Vol 303-306
◽
pp. 297-301
◽
2012 ◽
Vol 90
(9)
◽
pp. 1271-1280
◽