OASIS-P: Operable Adaptive Sparse Identification of Systems for fault Prognosis of chemical processes. (November 2021)

• Record Type: Journal Article
• Authors: Bhadriraju, Bhavana; Kwon, Joseph Sang-Il; Khan, Faisal
• Is Part Of: Journal of process control. Volume 107(2021)
• Page Start: 114
• Page End: 126
• Publication Date: 2021-11
• Subjects: Nonlinear systems -- Sparse model -- Neural networks -- Risk assessment -- Contribution plots -- Fault prediction -- Fault isolation; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2021.10.006
• Languages: English
• ISSNs: 0959-1524

Abstract

Abstract: With the increasing process complexities, data-driven fault prognosis has emerged as a promising fault management tool that predicts and manages abnormal events well in advance. In this paper, we develop a fault prognosis framework named 'OASIS-P' by integrating operable adaptive sparse identification of systems (OASIS), which is a data-driven adaptive modeling technique, with a risk-based process monitoring approach and contribution plots. Firstly, OASIS is employed with the risk assessment procedure for the prediction of impending faults. As the OASIS model is adaptive, it copes with the initial fault symptoms and forecasts the future behavior of the process under faulty conditions reasonably well, thereby providing an early fault prediction. Next, the fault isolation step is immediately initiated using contribution plots to identify the faulty variables. Unlike in fault diagnosis, the problem of ambiguity in interpreting contribution results due to fault propagation is not an issue in fault prognosis, if the fault isolation step is implemented at an early stage of the fault before it affects the other variables. Hence, the contribution plots together with OASIS can proactively monitor the process in real-time. As a case study, we demonstrate OASIS-P for fault prognosis of a reactor–separator system. Highlights: A data-driven fault prognosis framework is developed for chemical process systems. Sparse regression-based adaptive model is used to forecast process dynamics. Process risk is monitored to predict the occurrence of abnormal events. Contribution plots are used to proactively isolate faulty variables. Applicable for both single and simultaneously occurring independent faults.