PREDICTION OF THE COMPRESSIVE STRENGTH OF FOAM CONCRETE USING THE ARTIFICIAL NEURAL NETWORK

Abstract

Foam concrete experiments require a lot of time and money. Therefore, a brand new modeling system is needed. A system which is not depending on the experiments but can predict the strength of foam concrete accurately. In this study, an Artificial Neural Network (ANN) was used as a solution to predict the compressive strength of foam concrete. The ANN method uses the feed-forward backpropagation architecture and the Levenberg-Marquardt training algorithm it consists of three layers, namely the input layer, hidden layer, and output layer. The input layer consists of cement, sand, water, foam, slum flow, and density, while the output layer consists of the compressive strength of foam concrete. The number of data used in this study was 90 data. The results indicated that the Artificial Neural Network had 6 input layer neurons, 13 hidden layer neurons, and 1 output layer neurons and they had an accuracy rate of 98.7%. It can be concluded that the Artificial Neural Network method can be used to predict the strength of foam concrete with an accuracy level close to 100 percent.