Probabilistic Chloride Penetration Models and Corrosion Initiation Probability of RC Bridge Based on Long-Term Test Data

Abstract

Chloride penetration initiates the corrosion of the steel reinforcement and causes the deterioration of an RC bridge in a marine area. This study focuses on the probabilistic models in the chloride penetration process for the probabilistic corrosion initiation analysis of the steel reinforcement. The probabilistic model for the chloride diffusion coefficient is developed by adding correction terms to correct the potential bias in the existing deterministic model. The model parameters are calibrated and selected with the Bayesian approach based on long-term test data. The probabilistic models for the surface chloride concentration and chloride threshold are selected from reviews and discussions on the existing models. The probabilistic models developed or selected contain sufficient information on concrete quality and environmental condition influencing the anticorrosion performance of the RC structure. The corresponding probabilistic analysis approach is proposed with emphasis on the quantification for the influence of uncertainties presented in the model parameters. The time-variant probability curves of corrosion initiation for an example RC bridge are obtained using the proposed probabilistic models and analysis approach. The results show that the environmental condition has a great influence on the corrosion initiation reliability for steel reinforcement.