Capacidade resistente de vigas celulares mistas de aço de alta resistência e concreto à instabilidade no montante da alma

Abstract

This study aims to investigate the resistance capacity of high-strength steel composite cellular beams with concrete slabs against web-post buckling. To this end, nonlinear geometric and material analyses were performed using the ABAQUS® software. The numerical model was validated against experimental data from physical models of composite cellular beams with concrete slabs. Four reference models were used as a basis for a parametric study carried out in two stages. In the first stage, the influence of the steel yield strength was investigated for S460, S690, and S960 steels (with yield strengths of 460 MPa, 690 MPa, and 960 MPa, respectively), without geometric variations. In the second stage, variations in steel strength were combined with variations in the following ratios: the opening diameter to the unexpanded steel profile height (Do/d), taking values of 0.8, 0.9, 1.0, and 1.1; and the center-to-center spacing to opening diameter ratio (p/Do), taking values of 1.2, 1.3, 1.4, and 1.5. For the variations of the validated symmetric profile models, CCB1 from Nadjai et al. (2007) and CCB3 from Müller et al. (2006), as well as the asymmetric model CCB2 (asymmetry ratio of 1.13) also from Nadjai et al. (2007), web-post buckling remained the predominant failure mode across all steel grades during the first stage, with maximum strength gains ranging from 24.6% to 60.8% for S690 and from 54.0% to 83.4% for S960. For variations of model CCB4, also from Müller et al., (2006), which has a top-to-bottom tee asymmetry ratio of 2.8, only local web buckling of the top tee was observed in the first stage, regardless of yield strength, with strength gains of 28% for S690 and 68% for S960. In the second stage, increasing Do/d tended to reduce the maximum strength, while increasing p/Do generally increased it, although complex interactions between geometry and steel yield strength were observed. For the CCB1, CCB2, and CCB3 variations, as steel strength increased, failure modes shifted toward those involving the formation of Vierendeel plastic mechanisms, either in isolation or combined with web-post buckling. For the CCB4 variations, the failure mode shifted from pure local web buckling of the top tee to an interaction between this failure mode and the Vierendeel plastic mechanism. The numerically obtained maximum strength values were compared with four analytical verification models: Panedpojaman et al. (2014), Grilo et al. (2018), AISC Steel Design Guide 31 (2016), and BS EN 1993-1-13 (2024). The comparison shows that most formulations underestimate the maximum resistance capacity of high-strength steel–concrete composite cellular beams. When analyzing the effect of increasing steel yield strength on maximum strength, the AISC Steel Design Guide 31 (2016) method showed an increase in the F_numerical/F_analytical ratio, while the other methods showed a decrease in this ratio as strength increased. Furthermore, all methods exhibited high coefficients of variation, often exceeding 40%, indicating high relative data dispersion. This level of variability compromises the reliability of predictions for high-strength steels, suggesting that current formulations do not adequately capture the instability and plastification phenomena in composite cellular beams with high-strength steel profiles.