An on-line framework for monitoring nonlinear processes with multiple operating modes. (May 2020)

• Record Type: Journal Article
• Authors: Tan, Ruomu; Cong, Tian; Ottewill, James R.; Baranowski, Jerzy; Thornhill, Nina F.
• Is Part Of: Journal of process control. Volume 89(2020)
• Page Start: 119
• Page End: 130
• Publication Date: 2020-05
• Subjects: Fault detection -- Unsupervised learning -- Process monitoring -- Multimode process -- Kernel method; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2020.03.006
• Languages: English
• ISSNs: 0959-1524

Abstract

Highlights: This paper proposed an on-line framework for monitoring processes with multiple operating modes. The DP-GMM based clustering method and the NSDC-KPCA method are used for building monitoring models. Practical issues, including the anomaly indicator selection and the tuning of the NSDC kernel, are considered. The performance has been demonstrated using a numerical simulated dataset and an experimental benchmark dataset. Abstract: A multivariate statistical process monitoring scheme should be able to describe multimodal data. Multimodality typically arises in process data due to varying production regimes. Moreover, multimodality may influence how easy it is for process operators to interpret the monitoring results. To address these challenges, this paper proposes an on-line monitoring framework for anomaly detection where an anomaly may either indicate a fault occurring and developing in the process or the process moving to a new operating mode. The framework incorporates the Dirichlet process, which is an unsupervised clustering method, and kernel principal component analysis with a new kernel specialized for multimode data. A monitoring model is trained using the data obtained from several healthy operating modes. When on-line, if a new healthy operating mode is confirmed by an operator, the monitoring model is updated using data collected in the new mode. Implementation issues of this framework, including the parameter tuning for the kernel and the selection of anomaly indicators, are also discussed. A bivariate numerical simulation is used to demonstrate the performance of anomaly detection of the monitoring model. The ability of this framework in model updating and anomaly detection in new operating modes is shown on data from an industrial-scale process using the PRONTO benchmark dataset. The examples will also demonstrate the industrial applicability of the proposed framework.