Application of neural networks for estimation of concrete strength

Abstract

The compressive strength of concrete is mostly used criterion in producing concrete. However, testing for compressive strength of concrete specimens is a complicated and time-consuming task. More importantly, it is too late to make improvement if the test result does not satisfy the required strength, since the test is usually performed at the 28th day after the placement of concrete at the construction site. Therefore, strength prediction before the placement of concrete is highly desirable. This study presents the effort in applying the neural network technique for predicting the compressive strength of concrete based on concrete mix proportions. For training and testing, the data sets on the mix proportions of two ready mixed concrete companies were used, and then the required compressive strength was predicted by trial and error. The predicted compressive strengths were verified by comparing the predicted results with those tested in the laboratory. The results show that the neural networks are very efficient in predicting the compressive strength of concrete with good accuracy. The application of this technology in predicting the compressive strength of concrete is expected to contribute to the assurance of concrete quality for manufacturing of optimal concrete.