Novel Nonlinear Process Monitoring and Fault Diagnosis Method Based on KPCA-ICA and MSVMs

A novel nonlinear process monitoring method based on kernel principal component analysis (KPCA)-independent component analysis (ICA) and multiple support vector machines (MSVMs) is proposed. KPCA pretreats data and makes the data structure become as linearly separable as possible. ICA seeks the projection directions in the KPCA whitened space, making the distribution of the projected data as non-Gaussian as possible. MSVMs is applied for identification of different fault sources. The application to Tennessee Eastman process indicates that the proposed method can effectively capture the nonlinear relationship in process variables and has good diagnosis capability and overall diagnosis correctness rate.