ABSTRACT
When prior information about random variables is scarce, Bayesian inference method can reduce random variables uncertainty through integrating new information with prior information effectively. Based on Bayesian inference, we can obtain accurate reliability analysis results. Although Bayesian inference method has been widely used in some engineering fields (Rusk et al. 2011, Zhang & Mohammadian 2008, Zhang et al. 2009), few studies on its applications in mechanical time-dependent reliability area have been performed. A pin, the key component of a lifting mechanical system, is researched in this paper. Firstly, the elastic modulus and Poisson’s ratio of the pin are considered as random variables; secondly, the wear time-dependent reliability model of the pin is established based on Archard wear theory; thirdly, the probability density functions of elastic modulus and Poisson’s ratio are updated by Bayesian inference method in order to reduce the uncertainty; finally, based on PHI2 method, three examples about wear time-dependent reliability of the pin are analyzed: (i) updating elastic modulus only; (ii) updating Poisson’s ratio only; (iii) updating both elastic modulus and Poisson’s ratio.
