Stochastic constitutive model of structural steel based on random field of corrosion depth

In this study, a new method based on the random field model (RFM) of corrosion depth was developed to investigate the effect of corrosion randomness on the constitutive relationship of steel. Through Monte Carlo simulations of the main parameters of the RFM, a large number of corroded surfaces with different corrosion degrees corresponding to the general and offshore atmospheres were generated. Finite element parametric simulations were conducted to analyze the degradation laws and random characteristics of the mechanical properties of corroded steel. The results showed that the mechanical properties such as elastic modulus, yield strength and tensile strength all conformed to normal distribution. With an increase in the corrosion loss ratio, the mean values of all parameters decreased linearly (the yield plateau slope increased), and the standard deviations increased linearly. Finally, the probabilistic model of each mechanical property and a stochastic constitutive model of corroded steel were established.