Shear capacity evaluation of studs in steel-high strength concrete composite structures

Abstract

The shear capacity of stud shear connectors was the main parameter affecting the mechanical performance of steel-concrete composite structures. In this paper, three artificial neural networks (ANN) were developed to evaluate the shear capacity of stud shear connectors in steel-high strength concrete composite structures. The models was applied to high strength concrete, covering the compressive strength of concrete in 61.19 MPa∼200 MPa. Based on the correlation analysis, the main influential parameters, including the compressive strength of concrete, the diameter, height, yield strength, number, and pretension force of stud shear connectors, were selected as input variables to the models. The proposed models were trained and tested with 100-group test data gathered from previous studies. By comparing with existing empirical models, it was proved that the proposed Elman network and RBF network had high applicability and reliability for predicting the shear capacity of stud shear connectors in steel-high strength concrete composite structures. Subsequently the parametric sensitive analysis was carried out based on the BP network.