Performance evaluation of mechanical properties of self-compacting concrete using artificial neural network
This experimental study was undertaken to investigate the effects of using local materials (cement, fly ash, super-plasticizer, coarse aggre-gate, and sand) on the mechanical properties of Self-Compacting Concrete (SCC). For this purpose, a total of 31 mixtures of SCC were prepared by the neural network design methods. Furthermore, based on the experimental results, the neural network model-based clear for-mulations were developed to predict the mechanical properties of SCC. The test results have shown that mineral admixtures were very effective on hardened properties of SCC. In addition, it was found that the developed model by using neural network appeared to have a high predictive capacity of hardened properties of SCC with respect to regression and experimental.