Experimental investigation and comparative machine-learning prediction of compressive strength of recycled aggregate concrete

In this study, the idea of recycling the concrete wastes and reuse of them for reproduction of green concrete has been presented. Thus, we have tried to study mechanical parameters using recycled aggregate concrete. For this purpose, three mix designs including natural, recycled and recycled fiber concrete were tested. Moreover, at the end of the paper, estimation of compressive strength using ANN methods has been presented. Based on the results, the recycled concrete and recycled fiber concrete with the proposed mix design have a high compressive strength, and due to relatively high porosity of the recycled aggregate concrete, its density has decreased by 2.48% and its water absorption increased by 54% compared to the natural concrete. Two artificial intelligence methods of ANN and SVM benefit from a quite equal coefficient of consistency, and the results of 124 test specimens with the results obtained from SVM are in a better agreement. Finally, two artificial intelligence methods were compared with the MLR using K-fold cross-validation, indicating superior performance of the artificial intelligence.