Predicting the compressive strength and slump of high strength concrete using neural network

Abstract

High Strength Concrete (HSC) is defined as concrete that meets special combination of performance and uniformity requirements that cannot be achieved routinely using conventional constituents and normal mixing, placing, and curing procedures. HSC is a highly complex material, which makes modelling its behavior very difficult task. This paper aimed to show possible applicability of neural networks (NN) to predict the compressive strength and slump of HSC. A NN model is constructed, trained and tested using the available test data of 187 different concrete mix-designs of HSC gathered from the literature. The data used in NN model are arranged in a format of seven input parameters that cover the water to binder ratio, water content, fine aggregate ratio, fly ash content, air entraining agent, superplasticizer and silica fume replacement. The NN model, which performs in Matlab, predicts the compressive strength and slump values of HSC. The mean absolute percentage error was found to be less then 1,956,208% for compressive strength and 5,782,223% for slump values and R 2 values to be about 99.93% for compressive strength and 99.34% for slump values for the test set. The results showed that NNs have strong potential as a feasible tool for predicting compressive strength and slump values.