A novel dynamic PCA algorithm for dynamic data modeling and process monitoring

• Published in: Journal of process control Vol. 67; pp. 1 - 11
• Main Authors: Dong, Yining, Qin, S. Joe
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.07.2018
• Subjects: Dynamic data modeling; Dynamic PCA; Fault detection; Process monitoring; Process monitoring; Dynamic PCA; Fault detection; Dynamic data modeling

•The paper develops a novel dynamic PCA (DiPCA) algorithm to extract a set of dynamic latent variables that capture the most dynamic variations in the data with a given number of latent factors.•The new models generate a number of principal time series that are most correlated to their past and thus...