Enhanced dynamic approach to improve the detection of small-magnitude faults. (2nd June 2016)

• Record Type: Journal Article
• Authors: Bernal-de-Lázaro, J.M.; Llanes-Santiago, O.; Prieto-Moreno, A.; Knupp, D.C.; Silva-Neto, A.J.
• Is Part Of: Chemical engineering science. Volume 146(2016)
• Page Start: 166
• Page End: 179
• Publication Date: 2016-06-02
• Subjects: Fault detection -- Kernel methods -- Small-magnitude faults -- AUC measure -- Metaheuristic algorithms -- Latency time; Chemical engineering -- Periodicals; Génie chimique -- Périodiques; Chemical engineering; Periodicals; Electronic journals; 660
• DOI: 10.1016/j.ces.2016.02.038
• Languages: English
• ISSNs: 0009-2509

Abstract

Abstract: The conventional SPE and Hotelling׳s T 2 statistics may not work properly in the detection of incipient and small-magnitude faults. In this paper, an enhanced dynamic Multivariate Statistical Process Control approach is proposed, which combined with the dimension reduction techniques KPCA and KICA improved the detection of these types of faults. In the parameters choice task two metaheuristic algorithms were used. The kernel optimization criterion used involves the computation of the False Alarm Rate (FAR) and False Detection Rate (FDR) indicators, unified by the Area Under the ROC Curve (AUC). The proposal was tested with excellent results on the Tennessee Eastman (TE) process. Abstract : Highlights: A novel MSPC based on a EWMA approach with enhanced dynamic (EWMA-ED) is proposed. A proposal to improve the detection of small-magnitude faults is presented. The FAR and FDR indicators are unified by the Area under the ROC curve (AUC). The proposed approach reduced the latency times and the missing alarms. The unified indicator AUC is used as a kernel parameter optimization measure.