Bootstrap Confidence Interval on IOHMM Parameters for System Health Diagnostic Under Multiple Operating Conditions. Issue 2 (2020)

• Record Type: Journal Article
• Authors: Shahin, Kamrul Islam; Simon, Christophe; Weber, Philippe
• Is Part Of: IFAC-PapersOnLine. Volume 53:Issue 2(2020)
• Page Start: 10297
• Page End: 10303
• Publication Date: 2020
• Subjects: System health -- PHM -- Input Output Hidden Markov Model -- Condition based diagnostic -- Degradation design -- Condition monitoring; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2020.12.2764
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: The operating conditions have an important impact on system degradation. This paper uses the Input-Output Hidden Markov Model to represent the system degradation having multiple operating conditions. In this paper the bootstrap method is applied to estimate the model parameters and to diagnostic the system health. Parameters of the model are computed with 95% confidence intervals. The uncertainty about multiple data sequences and degradation speed is handled according to the operating conditions. A numerical application is given to explain the methodologies used to estimate the model parameters and diagnostic the system health.