An Improved FVS-KPCA Method of Fault Detection on TE Process

Abstract

For complex nonlinear systems of chemical industry process, traditional kernel principal component analysis (KPCA) methods are very difficult to calculate the kernel matrix for fault detection with large sample sets. So an improved fault detection method based on feature vector selection-KPCA (FVS-KPCA) is developed. This method can evidently reduce calculational complexity of fault detection and is applied to the benchmark of Tennessee Eastman (TE) processes. The simulation results show that the proposed method can effectively improve the speed of fault detection.