ANN Prediction of bond strength between steel rebar and concrete containing micro-silica, nano-silica and fibers

In this study, bond strength between steel rebar and concrete containing micro-silica, nano-silica and fibers are investigated. To this aimed, 36 cylindrical (10 cm*15 cm) specimens have been constructed using 12 different mix designs. ANN model was used in order to predict the experimental results. The applied model consists of six input parameters as micro-silica, nano-silica, fibers, aggregate/cement ratios, water/cement ratio and cement strength grades (325, 425 and 525 kg/cm2). The bond strength between steel rebar and concrete was also used as output parameter. This model is trained by experimental data for validation of experimental results predicted by the researchers of the data network is used for data processing. The predicted results were also validated by data from previous researches. The results indicate that the artificial neural network is a powerful tool for predicting the effects of various concrete admixtures on bond strength between steel rebar and concrete. Moreover, various effective design parameters should be considered in the predicting model which may potentially yield more precise results.