Industrial & Engineering Chemistry Research · 2005 · 321 citations · 11 references
ComputingChemical EngineeringComputer ScienceProcess ControlProcess Systems EngineeringTime Series AnalysisChange DetectionChange-point DetectionComputational Time Series AnalysisPrincipal Component AnalysisSystem IdentificationTime Series Feature ExtractionChemical Process SystemsCorrelationCorrelation MatrixFeature Drift DetectionMultivariate Process MonitoringProcess Drift MonitoringProcess MonitoringSignal Processing For Process Analysis
The authors propose a fast moving‑window PCA algorithm that recursively adapts the mean, variance, and correlation matrix and incorporates an N‑step‑ahead horizon for process monitoring. The method updates the PCA model within a sliding window by recomputing the decomposition after adjusting the mean, variance, and correlation matrix, and applies the model identified N steps earlier to analyze current observations. Experiments show the algorithm is faster than conventional moving‑window PCA when the window exceeds three times the number of variables, is insensitive to window size, and the N‑step horizon enhances detection of slow drifts in complex chemical processes.
This paper introduces a fast algorithm for moving window principal component analysis (MWPCA) which will adapt a principal component model. This incorporates the concept of recursive adaptation within a moving window to (i) adapt the mean and variance of the process variables, (ii) adapt the correlation matrix, and (iii) adjust the PCA model by recomputing the decomposition. This paper shows that the new algorithm is computationally faster than conventional moving window techniques, if the window size exceeds 3 times the number of variables, and is not affected by the window size. A further contribution is the introduction of an N-step-ahead horizon into the process monitoring. This implies that the PCA model, identified N-steps earlier, is used to analyze the current observation. For monitoring complex chemical systems, this work shows that the use of the horizon improves the ability to detect slowly developing drifts.
11
Recursive PCA for adaptive process monitoring
Weihua Li, Heng Yue, Sergio Valle‐Cervantes et al. · Journal of Process Control · 2000 · 844 citations
Recursive PLS algorithms for adaptive data modeling
S. Joe Qin · Computers & Chemical Engineering · 1998 · 596 citations