Application of recursive statistical-based techniques for industrial process monitoring

• DOI: 10.1109/icosp.2004.1442330
• Publication Date: 2005-06-28T14:42:50Z
• Authors: Xun Wang; U. Kruger; Pingkang Li; Yong Wang
• Keywords: Statistical Process Control; Sliding window protocol; Fluid catalytic cracking

This paper discusses the monitoring of large-scale processes that exhibit non-stationary and/or time-varying behavior. The work considers statistically-based techniques relate to multivariate statistical process control (MSPC) framework. Particular focus receives partial least squares (PLS), as it allows distinguishing between cause and effect variables. To demonstrate difficulties show behavior, use conventional PLS is contrasted with its recursive moving-window counterparts. An application study a realistic simulation fluid catalytic cracking unit (FCCU) (i) shows behavior (ii) raises attention on how apply moving window correctly.