Performance modeling for condition-based activation of the redundant safety system subject to harmful tests

Aibo Zhang, Songhua Hao*, Peng Li, Min Xie, Yiliu Liu

*Corresponding author for this work

    Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

    5 Citations (Scopus)

    Abstract

    Redundancy is a widely adopted measure for the enhancement of safety system performance. This fault-tolerance design can keep the system functioning through one unit if the other has failed. Besides the continuous aging degradation, mechanical units expose to random shocks, resulting in the dependence. In practice, it is often sufficient to activate only one unit in case of a shock so that the degradation will then be intensified. Unit-level failure occurs when the total degradation level reaches a certain level and will keep hidden. Thus, periodic tests are arranged to reveal the system state, accompanying the side effects on the existing degradation given the test-induced additional stress. A novel condition-based activation policy is proposed based on the system state, referring to how to preset the allocation of units to withstand potential shocks in the upcoming test interval. In this paper, analytical formulas are developed to evaluate system performance with the involvement of these aforementioned factors. Finally, numerical examples are presented to find an optimal test and activation policy to reach minimum system unavailability in a given service time horizon. This study is expected to provide clues for practitioners in the optimal test and operation of redundant structure.
    Original languageEnglish
    Article number108649
    JournalReliability Engineering and System Safety
    Volume226
    Online published17 Jun 2022
    DOIs
    Publication statusPublished - Oct 2022

    Funding

    This work is supported by National Natural Science Foundation of China (71971181, 72032005 and 72101170) and by Research Grant Council of Hong Kong (11203519 and 11200621). It is also funded by Hong Kong Innovation and Technology Commission (InnoHK Project CIMDA) and Hong Kong Institute of Data Science (Project 9360163), the IKT PLUSS program of Norwegian Research Council (Project number 309628) and Sichuan Science and Technology Program 2021YFH0068.

    Research Keywords

    • Activation policy
    • Degradation performance
    • Harmful tests
    • Optimal test
    • Redundant structure
    • Safety systems

    Fingerprint

    Dive into the research topics of 'Performance modeling for condition-based activation of the redundant safety system subject to harmful tests'. Together they form a unique fingerprint.

    Cite this