Pore structure and prediction of mechanical properties by ANN of concrete mixed MK and SF under hydrochloric acid corrosion

Abstract

It is necessary to evaluate the acid resistance of concrete to ensure the safety. In this study, the effect of 5% metakaolin (MK) and silica fume (SF) with 5%, 10%, or 15% dosages on hydrochloric acid resistance of concrete was investigated. Mechanical properties including axial compressive strength, flexural strength, and splitting tensile strength were studied. Besides, the pore structure of concrete was analyzed by mercury intrusion porosimetry (MIP) according to the porosity, pore connectivity, pore size distribution, and pore volume distribution. The results showed that 5% MK and 5% SF had better acid resistance for concrete. In addition, an artificial neural network (ANN) model was proposed to predict the axial compressive strength, flexural strength, and splitting tensile strength of concrete containing MK and SF under hydrochloric acid corrosion. The mass (cement, MK, and SF) and the acid corrosion days were input parameters in the ANN model. It was found that the average relative error of the developed ANN model was 6.8%, indicating that the ANN model had good potential in predicting the mechanical properties of concrete containing MK and SF under hydrochloric acid corrosion. This study can provide the guide on the improvement the acid resistance of concrete.