高技术通讯（英文版）2005，Vol.11Issue(2)：189-192,4.
Nonlinear fault diagnosis method based on kernel principal component analysis
Nonlinear fault diagnosis method based on kernel principal component analysis
摘要
Abstract
To ensure the system run under working order, detection and diagnosis of faults play an important role in industrial process. This paper proposed a nonlinear fault diagnosis method based on kernel principal component analysis (KPCA). In proposed method, using essential information of nonlinear system extracted by KPCA, we constructed KPCA model of nonlinear system under normal working condition. Then new data were projected onto the KPCA model. When new data are incompatible with the KPCA model, it can be concluded that the nonlinear system isout of normal working condition. Proposed method was applied to fault diagnosison rolling bearings. Simulation results show proposed method provides an effective method for fault detection and diagnosis of nonlinear system.关键词
kernel principal component analysis/fault diagnosis/nonlinearKey words
kernel principal component analysis/fault diagnosis/nonlinear引用本文复制引用
Yan Weiwu,Zhang Chunkai,Shao Huihe..Nonlinear fault diagnosis method based on kernel principal component analysis[J].高技术通讯（英文版）,2005,11(2):189-192,4.基金项目
Supported by the High Technology Research and Development Programme of China (No. 2001AA413130). （No. 2001AA413130）
