Tensile strength prediction method through compressive concrete cube test

The tensile strength of concrete is a crucial factor in civil engineering design of buildings and structures. Engineers are trying to find a new way to calculate tensile strength by using different characteristics in terms of the mechanical properties of concrete. This work aims to find the tensile strength through the apparent failure of cubic compression samples. In this way, the fracture angle, the diameter of the cubic sample, and the compressive strength were the factors most important in the tensile strength. Method. Two types of concrete (shredded and minibar high-performance concrete) have been chosen to find the equation of tensile strength. The tensile strength equation was obtained according to the logistic algorithm and micro tensile and compressive stresses, which finally shows that the equation is related to the compressive strength and crack behavior of the cubic compressive sample. Moreover, to find the accuracy of machine learning models, correlation coefficient (R²), mean absolute errors (MAE), and root mean square error (RMSE) are generated. Results. The results show that the experimental and prediction results are close together. For example, R², MAE, and RMSE were 0.93, 0.023, and 0.00093 for experimental results and 0.99, 0.083, and 0.00014 for prediction results, respectively. It should be noted that the equations and display methods in this research can find the tensile strength with high accuracy.