Fault Diagnosis Method on Polyvinyl Chloride Polymerization Process Based on Dynamic Kernel Principal Component and Fisher Discriminant Analysis Method

Abstract

In view of the fact that the production process of Polyvinyl chloride (PVC) polymerization has more fault types and its type is complex, a fault diagnosis algorithm based on the hybrid Dynamic Kernel Principal Component Analysis-Fisher Discriminant Analysis (DKPCA-FDA) method is proposed in this paper. Kernel principal component analysis and Dynamic Kernel Principal Component Analysis are used for fault diagnosis of Polyvinyl chloride (PVC) polymerization process, while Fisher Discriminant Analysis (FDA) method was adopted to make failure data for further separation. The simulation results show that the Dynamic Kernel Principal Component Analyses to fault diagnosis of Polyvinyl chloride (PVC) polymerization process have better diagnostic accuracy, the Fisher Discriminant Analysis (FDA) can further realize the fault isolation, and the actual fault in the process of Polyvinyl chloride (PVC) polymerization production can be monitored by Dynamic Kernel Principal Component Analysis.

A Novel Process Monitoring Method Based on Dynamic Related ReliefF-SFA Method

“…(4) Determine the control limit by Equation (12) and (13) Online Monitoring: (1) Collect the i th i(i ≥ q) moment monitoring data, building the online sampling vector according Equation (19) and new standardized data y new is obtained.…”

“…These algorithms improve the fault detection performance of SFA, however, they cannot solve the problem of the high dimension and high correlation between data variables. Although such as dynamic SFA [17], dynamic PCA [18], and dynamic fisher discriminant analysis (DFDA) [19] consider the cross-correlation and auto-correlation between variables, but Kruger et al [20] proved that the traditional dynamic method cannot completely eliminate the autocorrelation of process variables, especially in some cases with strong autocorrelation, the features obtained by using the traditional dynamic method still have autocorrelation and crosscorrelation question. Especially in industrial process, most variables have a strong correlation, hence it is still difficult to apply them to fault diagnosis of the real systems with strong autocorrelation.…”

Fault detection and diagnosis of nonlinear dynamical processes through correlation dimension and fractal analysis based dynamic kernel PCA

“…DPCA extension to nonlinear processes through Dynamic Kernel PCA (DKPCA) was firstly proposed in [36], integrating the potentials of KPCA and the dynamical properties of DPCA. DKPCA has been modified to batch DKPCA in [37], and incorporated with FDA (DKPCA-FDA) in [38] for fault isolation. These methods aim at capturing nonlinear static dependencies, however, their dynamical structure and the number of PCs are still identified using linear approximations that fail to address nonlinear dynamics.…”

An Evaluation Method for Sortie Generation Capacity of Carrier Aircrafts with Principal Component Reduction and Catastrophe Progression Method

“…However, these evaluation methods are one-sided and subjective, which ignore the correlation and contradiction of indexes. AHP is difficult to evaluate the multivariate evaluation objects objectively [18][19][20][21][22][23][24][25][26].…”