Recycled Aggregates Concrete Compressive Strength Prediction Using Artificial Neural Networks (ANNs)

The recycled aggregate is an alternative with great potential to replace the conventional concrete alongside other benefits such as minimising usage of natural resources in exploitation produce new concrete. Eventually, this will lead reducing construction waste, carbon footprints and energy consumption. This paper aims study compressive strength using Artificial Neural Network (ANN) which has been proven be a powerful tool for use predicting mechanical properties Three different ANN models where 1 hidden layer 50 number neurons, 2 layers (50 10) neurons (modified activation function) (60 3) are constructed aid Levenberg-Marquardt (LM) algorithm, trained tested 1030 datasets collected from related literature. 8 input parameters cement, blast furnace slag, fly ash, water, superplasticizer, coarse aggregate, fine age used training models. layers, type algorithm affect prediction accuracy. predicted aggregates shows compositions admixtures binders, water–cement ratio furnace–fly ash greatly properties. results show that predictable very high accuracy proposed ANN-based model. model can further optimising proportion waste material ingredients targets strength.