NEURAL-NETWORK MODELING OF CHLORIDE BINDING

The capacity of cement to bind chloride is considered to be one of the controlling parameters in the process of chloride-induced corrosion o f steel in concrete. In this work, a literature review has been undert aken to identify, factors affecting chloride binding. Data from 21 pre viously published works have been collated and used to develop a neura l network model to predict the free chloride concentration as a functi on of 18 input variables. These predictions are relatively free of the indiscriminate variations present in individual measurements. The inf luence of factors not yet fully evaluated is identified In addition, t he relative importance of a wide range of factors is quantified. The n eural network model provides a powerful tool for generating binding is otherms for use in models of chloride ingress into concrete.