An intelligent model for the prediction of the compressive strength of cementitious composites with ground granulated blast furnace slag based on ultrasonic pulse velocity measurements

In this study, the compressive strength of cementitious composite containing ground granulated blast furnace slag (GGBFS) has been predicted. For this purpose, the intelligent models: the self-organizing feature map (SOFM) and the artificial neural network (ANN) were used and compared. A database containing 84 sets of data was created based on the time and mixture proportions of concrete, as well as on nondestructive ultrasonic pulse velocity measurements. It was proved that the developed model of predicting the compressive strength of the green cementitious composites containing GGBFS was accurate. It was also simple as it contained only three parameters that were used as input variables. The novelty of this research is the fact that they can be performed on existing structures, not only after 28 days, but also at early ages (3 and 7 days) and much longer periods (after 150 and 180 days). This makes this method more universal and increases the possibility of it being useful for construction practice.

Automated summary

In this study, the compressive strength of cementitious composite containing ground granulated blast furnace slag (GGBFS) was accurately predicted using intelligent models. The novelty of this research lies in its wide applicability, as it can be used to test structures at different time periods.