Title :
An improved Kernel method for fault diagnosis
Author :
Cui, I.F. ; Guo, G.S. ; Miao, M.X. ; Liu, S.X.
Author_Institution :
Dept. of Mech. & Electr. Eng., Zhengzhou Inst. of Aeronutical Ind. Manage., Zhengzhou
Abstract :
Kernel Fisher discriminant analysis (KFDA) has been widely used in fault diagnosis. In this paper, a feature vector selection (FVS) scheme based on a geometrical consideration is given to reduce the computational complexity of KFDA when the number of samples becomes large. Experimental results show the effectiveness of our method.
Keywords :
computational complexity; fault diagnosis; manufacturing processes; statistical analysis; computational complexity; fault diagnosis; feature vector selection scheme; improved Kernel method; kernel Fisher discriminant analysis; Computational complexity; Engineering management; Fault diagnosis; Feature extraction; Independent component analysis; Kernel; Manufacturing processes; Principal component analysis; Scattering; Vectors;
Conference_Titel :
Systems and Control in Aerospace and Astronautics, 2008. ISSCAA 2008. 2nd International Symposium on
Conference_Location :
Shenzhen
Print_ISBN :
978-1-4244-3908-9
Electronic_ISBN :
978-1-4244-2386-6
DOI :
10.1109/ISSCAA.2008.4776167
