Diagnosis based on improved principle component analysis
Jiang Li-ying · 2009
A kind of fault diagnosis method based on improved PCA and the fuzzy clustering is proposed,which use shorter diagnosis time and higher accuracy.We extract fault feature subset by PCA,then calculate values of T2 and Q,if they surpassed the threshold,identify the fault class using the fuzzy clustering,The application of Tennessee-Eastman Process(TEP) shows the validity of the proposed method.