Artificial neural networks to assess the useful life of reinforced concrete elements deteriorated by accelerated chloride tests

In order to analyse the behaviour of concrete exposed to chloride attack, 243 specimens of 5 cm in diameter and 10 cm in thickness were prepared to analyse the influence of water/cement ratio, mineral additions, type of cement, period of curing and level of exposure on the penetration of chloride ions. The aim of this paper is to get chloride depth penetration and chloride diffusion of concrete specimens under conditions of drying–wetting cycles. Based on the experimental results, an Artificial Neural Network (ANN) modelling was used to map the relationship between the variables analysed and the ion penetration depth. Results obtained showed that ANN modelling proved to be efficient to estimate the depth of chloride penetration and chloride diffusion coefficients in concrete, and the parameters that most influence the depth of chloride penetration were the type of cement, the type of addition and the cure time. •Reinforced concrete elements deteriorated by accelerated chloride tests.•Chloride penetration depth assess by Artificial Neural Networks (ANN).•Chloride diffusion coefficient assess by Artificial Neural Networks (ANN).•Experimental campaign to study chloride penetration depth and chloride diffusion coefficient of different types of concretes.