Reliability assessment on the degradation properties of polymers under operating temperature and vibration conditions

Abstract

This study focuses on the design of polymer components considering their degradation under designed operating conditions in automobiles. We use stochastic and statistical methods to ensure that such components are reliable and robust. The behaviours of polymers are described using a viscoelastic model, and degradation properties of polymers are obtained from creep and tensile data that are acquired at various temperatures. Using the Maxwell fluid model, we calculate the Prony series, which estimates viscoelastic models based on creep data. By considering Prony coefficients that describe degradation characteristics, this approach generates stress data via a frequency-response analysis of polymer components in automobiles. These data are used to generate performance functions by the response surface method. We assess the reliability considering the variation of temperature-dependent degradation properties and the areas of the peak frequency. In this study, degraded properties and frequencies are assumed to have a normal distribution, and we evaluate the reliability and probability of failure under the yield strength criteria using a Monte Carlo simulation. We then compare the reliability and failure probabilities of the given polymers in an automotive component. Based on these comparisons, we suggest the most suitable polymeric materials for use in automotive applications.