Prediction of corrosion behaviour of reinforced concrete with manufactured sand using artificial neural network

Abstract

The present study focuses on evaluating the corrosion performance characteristics of steel embedded in concrete, in which Manufactured Sand (M-Sand) is utilized as a fractional and full replacement for natural sand. Corrosion characteristics of steel that is fixed in concrete are analysed by accelerated corrosion test for M40 grade concrete supplanted with 60% manufactured sand for river sand which is found optimum. The modelling of corrosion currents is done using feed forward Artificial Neural Network. The test results exhibits that the durability property of concrete against the effect of corrosion is enhanced, by the partial substitution of 60 % M-Sand. Accurate modelling results for corrosion currents was obtained using Artificial Neural Network (ANN). The test results exhibits that ANN structures produced close prediction current values to that of experimental results.