Artificial intelligence approach for predicting compressive strength of foamed concrete

Abstract

Accurately predicting the compressive strength of foamed concrete plays a key role in the wide application of foamed concrete in practice. This study investigates the performance of the six AI models in estimating the compressive strength of foamed concrete. A dataset of 150 samples available in the literature was used for training and testing the AI models. The dry density, cement and sand content, and water-to-cement ratio were employed as input parameters, while the 28-day compressive strength was used as the output parameter. Four statistical indicators were utilized to evaluate the performance of the AI models. The study results reveal that the AI models yield an accurate prediction of the compressive strength of foamed concrete. The best performance model in estimating the compressive strength of foamed concrete is the M5Rules model, while the least accurate model depends on the indicators used to measure the accuracy of the AI models.

Keywords

• compressive strength of foamed concrete
• artificial intelligence models
• prediction of the compressive strength of foamed concrete