Fault Diagnosis Method of Offset Printer Feeding Mechanism Based on Kernel Principal Component Analysis and K-Means Clustering
暂无分享,去创建一个
The printing machine is a sort of large-scale equipment characterized by high speed and precision. A fault diagnosis method based on kernel principal component analysis (KPCA) and K-means clustering is developed to classify the types of feeding fault. The multidimensional and nonlinear data of printed image could be reduced by KPCA to make up the deficiency of the traditional K-means clustering method. In this paper, it is experimentally verified that the classification accuracy of the combined method is higher than the traditional clustering analysis method in feeding fault detection and diagnosis. This method provides a shortcut for the determination of fault sources and realizes multi-faults diagnosis of printing machinery efficiently
[1] Bernhard Schölkopf,et al. Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.
[2] Jie Zhang,et al. Performance monitoring of processes with multiple operating modes through multiple PLS models , 2006 .
[3] Masoud Soroush,et al. A method of sensor fault detection and identification , 2005 .