T-S fuzzy dynamic fault tree analysis based on expert comprehensive assessment

The T-S dynamic fault tree (DFT) effectively models the dynamic failure behaviour of systems through the application of the T-S dynamic gate rule. However, obtaining failure data for components within high-reliability systems often presents a significant challenge. To enhance the T-S DFT's capability to address uncertain events, this paper introduces the T-S fuzzy dynamic fault tree (FDFT) analysis method, grounded in comprehensive expert assessments. This approach aggregates the evaluations from various experts regarding the likelihood of each fault event within the system and converts these assessments into quantitative interval fuzzy data. Subsequently, these interval fuzzy numbers are incorporated into the T-S dynamic gate input and output rule algorithm of the T-S DFT. This integration facilitates the determination of the system’s failure occurrence probability over time, as well as the probabilistic importance and critical importance of each failure cause. The feasibility of this method is demonstrated through a case study involving the bogie-bearing system of a moving train set. This method provides a robust foundation for fault analysis and maintenance of high-reliability systems where fault data are scarce.