Back to Search
Start Over
Improved Kernel Principal Component Analysis and Its Application for Fault Detection.
- Source :
- Advanced Intelligent Computing Theories & Applications. With Aspects of Theoretical & Methodological Issues (9783540874409); 2008, p688-695, 8p
- Publication Year :
- 2008
-
Abstract
- The kernel principal component analysis (KPCA) based on feature vector selection (FVS) is proposed in this paper for fault detection in nonlinear system. Firstly, the KPCA algorithm is described in detail. Secondly, a feature vector selection (FVS) scheme based on a geometric consideration is adopted to reduce the computational cost of KPCA. Finally, the KPCA and KPCA based on FVS (FVS-KPCA) are applied to a simple nonlinear system. The fault detection results and the comparison confirm the superiority of FVS-KPCA in fault detection. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISBNs :
- 9783540874409
- Database :
- Complementary Index
- Journal :
- Advanced Intelligent Computing Theories & Applications. With Aspects of Theoretical & Methodological Issues (9783540874409)
- Publication Type :
- Book
- Accession number :
- 76725802
- Full Text :
- https://doi.org/10.1007/978-3-540-87442-3_85