Continuous-time T-S Dynamic Fault Tree Analysis Method

Compared with DUGAN dynamic fault tree analysis method, discrete-time T-S dynamic fault tree analysis method enhances the ability of the fault tree to describe static and dynamic failure behavior, and can directly analyze a system directly quantitatively, but it has analysis and calculation errors, and can not reflect the change trend of system reliability. To this end, continuous-time T-S dynamic fault tree analysis method is proposed. The continuous-time T-S dynamic gates describing static and dynamic failure behavior are defined, a method of constructing continuous-time T-S dynamic gates describing rules based on the temporal relationship of lower-level events and the description of events occurrence impulse function and the possibility of higher-level event occurrence is proposed by using the integral characteristic of impulse function at impulse point. Then analysis and calculation methods based on the description of the rule execution degree and the possibility of the higher-event and the higher-event failure probability density function and the failure probability distribution function of the impulse function integration are proposed. The proposed continuous-time T-S dynamic fault tree analysis method has stronger failure behavior description ability than DUGAN dynamic fault tree analysis method, it solves the problem of analysis and calculation errors of discrete-time T-S dynamic fault tree analysis method and can reflect the change trend of system failure probability. Compared with Markov chain, discrete-time Bayesian network, continuous-time Bayesian network solving DUGAN dynamic fault tree and discrete-time T-S dynamic fault tree analysis method, the feasibility of continuous-time T-S dynamic fault tree analysis method is verified and continuous-time T-S dynamic fault tree analysis method has more advantages. © 2020 Journal of Mechanical Engineering.