Service Life Prediction Of Concrete Structures Under Chloride Environment Using Stochastic Finite Element Method For Reliability Analysis

Chloride-induced corrosion of reinforcing steel in concrete is one of the main reason which reduces the durability of concrete structures in chloride environment. As the material property and environment factors are time-dependent and random, it is need to establish a probability method for prediction of durability service life of concrete structures under chloride environment. This paper developes a perturbation stochastic finite element method (SFEM) for analysis of chloride diffusion in concrete, incorporing the time-dependence into the mathematical model of the chloride diffusion coefficient and the surface chloride concentration, and presents a probability method for service life evaluation based on reliability theory.The thesis includes the following conclusions:1. Based on the diffusion equation of chloride ion for inhomogeneous media, the local average method is employed to discretize the random field. According to the variational principle and the perturbation technique, the governing equations for the SFEM were established, by which the mean and standard deviation of chloride concentration can be determined. The SFEM is not sensitive to the correlation structure of random field and the size of time step. With the increases of coefficient of variation of the random field, the mean of the chloride ion concentration is almost unchanged, while the coefficient of variation of that increases gradually.2. This paper develops a perturbation stochastic finite element method (SFEM) for analysis of chloride diffusion in concrete, incorporing the time-dependence into the mathematical model of the chloride diffusion coefficient and the surface chloride concentration, in which the chloride ion diffusion coefficient of concrete is modeled as a Gaussian random field. The numerical results demonstrate the efficiency and accuracy of the presented method. It can be concluded that the randomness and the time dependence of the chloride ion diffusion coefficient significantly influence the process of chloride diffusion in concrete, while the effect of time-dependence of surface chloride concentration can be neglected for concrete structures exposed to chloride environment for a long-term duration.3. A probability method based on the reliability index for service life prediction of concrete structures is presented, incorporing the time-dependence and randomness to the chloride diffusion coefficient. The durability of concrete structures will be overestimated without considering the randomness of the chloride ion diffusion coefficient, while regardless of the time-dependence of the chloride ion diffusion coefficient will result in underestimating the service life of concrete structures. 4. The inverse error function of the analytical model for chloride ion diffusion is fitted, and an analytical fitting model for predicting the service life of concrete structures is determined. A fast prediction method for the service life of concrete structures is proposed by fitting a chloride concentration—time calculation model, and which can more efficiently estimate the service life of the existing concrete structure.