ABSTRACT
According to the design codes, the concrete cover is simply selected based on the predefined exposure classes. This implies a thicker concrete cover if moisture or chloride is expected to attack the concrete. However, a thicker concrete cover will result in larger cracks assuming the same area of reinforcement and dimensions of the reinforced concrete (RC) structure, or the area of reinforcement has to be increased. Considering chloride ingress as observed in marine environment, the ion transport is significantly affected by the crack pattern, which is characterized by a remarkable degree of uncertainty. In the scope of this work, by considering aleatory and epistemic uncertainties, an optimal concrete cover is determined using a multiphysics finite element model (FE). Since usually a large number of realization is needed, which will result in high computation time of a fully coupled multiphysical FE model, two subproblems are solved. The first FE model simulates tensile cracking of a concrete beam under mechanical loading, and the second FE model computes the coupled moisture and chloride transport within the cracked unsaturated concrete. Additionally, the enhanced diffusion of cracked concrete is accounted for by adjusting the diffusion coefficients of moisture and chloride. In this concept, the uncertainty of the design parameter (concrete cover) is quantified by an interval. This uncertain parameter and uncertain structural actions lead to polymorphic uncertain prognoses of cracks and subsequently to a polymorphic uncertain prediction of corrosion initiation. In conclusion, within the framework of this work, the crack induced corrosion initiation time will be maximized by optimizing the concrete cover.
