A Stochastic Computational Framework to Investigate the Initial Stage of Corrosion in Reinforced Concrete Superstructures

This article presents a stochastic computational framework to investigate the chloride-induced corrosion of reinforced concrete superstructures. For this purpose, three-dimensional finite-element models are developed to determine the extent of chloride penetration into the superstructure components. One of the unique capabilities of this framework is to simultaneously consider all the major factors that affect the corrosion process. Furthermore, the developed framework integrates various sources of uncertainty into the performance predictions. This will be achieved by modifying the element properties and boundary conditions of the finite-element models at each time step. For a reliable durability assessment of deteriorating structural components, the proposed framework incorporates the temporal and spatial uncertainties of the influential parameters into the finite-element analysis. Considering that most of the parameters involved in the corrosion process follow non-normal distributions, a series of non-Gaussian stochastic fields are generated following a computationally efficient procedure. The results calculated from extensive stochastic simulations are expressed in terms of the likelihood and extent of corrosion initiation. The outcome of this study indicates that the computational investigation of the corrosion process can be significantly improved if a reliable probabilistic framework is employed instead of conventional deterministic approaches.