Dynamic reliability prediction of vehicular suspension structure with damping random uncertainties

Abstract

This research is conducted such that a two degrees of freedom nonlinear stochastic vibration model of vehicular suspension structure with random bilinear damping forces and cubic nonlinear restoring forces is established. Based on extreme value theory (EVT) and Monte Carlo integration method (MCIM), a novel method of predicting the failure probabilities of the maximum amplitude responses of suspension structure is proposed. The extreme value distribution functions of amplitude responses, which are analytical expressed by underlying amplitude distribution functions, are acquired by using EVT. In order to obtain the analytical expressions of the underlying distribution functions, histograms are used to estimate their distribution forms. Under the log-normal distribution assumption, the means and standard deviations of the underlying distributions are calculated by using MCIM, and then by setting up amplitude failure levels, the extreme distribution functions and failure probabilities of amplitude responses are obtained. The Monte Carlo simulation method-based numerical analyses give enormous supports to use of the proposed prediction means.

Reliability-based assessment of reasonable mid-tower lateral stiffness for a three-tower suspension bridge

“…Matteo et al (1994) combined the Monte Carlo method and extreme value theory to evaluate the safety factor of the suspension bridge main cable. Wang et al (2019) also developed a novel approach to predict the dynamic reliability of vehicular suspension structures based on extreme value theory and the Monte Carlo integration method. Li et al (2013) used Latin hypercube sampling (LHS) to analyze the sensitivity and reliability of a self-anchored suspension bridge.…”