A key performance indicator-based fault detection scheme for marine diesel turbocharging systems. Issue 17 (November 2021)

• Record Type: Journal Article
• Authors: He, Zhichen; Yang, Ying; Han, Huayun; Wang, Jia; Zhang, Yuhan; Li, He
• Is Part Of: Journal of the Franklin Institute. Volume 358:Issue 17(2021)
• Page Start: 9346
• Page End: 9363
• Publication Date: 2021-11
• Subjects: Science -- Periodicals; Technology -- Periodicals; Patents -- United States -- Periodicals; 505
• DOI: 10.1016/j.jfranklin.2021.09.024
• Languages: English
• ISSNs: 0016-0032

Abstract

Abstract: In this paper, a thermodynamic mechanism-based key performance indicator (KPI) construction is proposed for the diesel engine turbocharging system, upon which the corresponding fault detection scheme is studied. To be specific, the efficiency of the turbocharger is first selected as the KPI and constructed by the thermodynamic mechanism. Based on this, the relation between the KPI and the measurable correlated variables is built as the KPI-related model. Then, the subspace identification approach is applied to identify the established model. By applying T 2 test statistic on the residual of the KPI-related model, a fault detection scheme is presented for system monitoring. Finally, the effectiveness of the proposed scheme is demonstrated on a simulation of the real turbocharger in the large ocean-going bulk carrier.