Abstract

Accurate and reliable predictions of the shear resistance of welded headed studs in deck slab ribs transverse to supporting beams are essential for achieving safe and economical designs of steel–concrete composite floors. This paper presents the development of a new empirical design model for computing the stud resistance, based on machine learning methods combined with regression models, and its calibration via reliability analyses according to the Eurocodes and US design practice. As an alternative to using the new equations directly, reduction factors have also been developed from the proposed model, which can be applied to three existing resistance models for studs in solid slabs. Comparisons of the proposed model with seven existing models indicate that the former outperformed the latter. At the same time, the proposed model produces higher shear resistances while satisfying the reliability requirements for many combinations of parameters, especially for lower concrete strengths common in composite floors, thicker decks, and through-hole stud welding. It was also found that the proposed model is insensitive to the following parameters considered important by the existing models: the concrete elastic modulus, the normal force on the slab face (ranging from 0 to 20% of the longitudinal shear force), stud welding type (through-deck or holes), location of slab reinforcement, and deck yield strength. The proposed model produces shear stud predictions that can be confidently used in design whilst providing considerably higher resistances for certain parameter combinations than those computed with the existing models, including those adopted by Eurocode 4 and AISC 360.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call