Ultimate Capacity Prediction of Concrete Slabs Reinforced with FRP Bars

In this paper, a relationship is presented to estimate the ultimate capacity of the punching shear in concrete slabs reinforced with FRP bars. For this purpose, neural networks as well as laboratory data are used. In the modeling, an input vector including geometric parameters of the section and the details of FRP are considered to determine the output of the system. The proposed relationship is a combination of a linear polynomial equation between inputs and a sigmoid function, which together can be able to estimate the selective output. To increase the accuracy of the system, all amounts of the inputs and output variables of the system were placed in a range between -1 to +1. The final proposed equation in this paper has a correlation coefficient of 0.94 for the tuning step of the equation coefficients. Also, the correlation coefficient for the evaluation phase during training and also for the test phase of the model was 0.98 and 0.97, respectively.

Keywords:

FRP , Concrete , Shear strength , Slab