PSI - Issue 84

Gian Felice Giaccu et al. / Procedia Structural Integrity 84 (2026) 991–998

993

elements along the interface. Two types of connectors are considered: rigid links for full composite action, and flexible links for partial interaction.

(a) (c) Fig. 1. Investigated generalized bridge deck: (a) cross-section and plane view for cable located in the boundary deck position (T1 Case), (b) cross-section and plane view for cable located in the inner deck position (T2 Case) and (c) schematic of the examined cross-sections of the cable stayed bridge. The longitudinal stiffness of flexible connectors is assigned based on literature recommendations (Kalibhat & Upadhyay 2020; Oehlers & Bradford 1995), with a nominal stiffness value of = 1×10 5 N/mm 2 adopted following Zeng et al. (2019) and AASHTO (2017). To ensure axial loading conditions with minimal bending influence, axial forces were applied horizontally at the steel girder ends, while vertical displacements were restrained at the centroid of the concrete slab. This setup effectively suppressed sagging or hogging deformations, isolating the axial–flexural interaction of interest. The mesh refinement was guided by nonlinear convergence criteria, with iterative adjustments performed until stress differences between successive iterations were below 2%. The Max Stress failure criterion was used in conjunction with a nonlinear elastic stress–strain formulation (Shuguang 2020). Geometric variability was introduced by modifying the deck's aspect ratio to account for slab asymmetry. The parametric matrix comprises 52 distinct numerical models, covering a range of concrete strengths (Rck 30 and 50 N/mm²) and varying mechanical area ratios between the concrete slab and steel girders. (b)

(a) (b) Fig. 2. Linear and non-linear Finite Element models: (a) with cables at edges span – T1 Case and (b) with cables at mid-span – T2 Case. 2.1. Linear analysis The parametric investigation evaluates the generalized composite bridge deck by varying the aspect ratio parameter  (0.06, 0.18, and 0.24), considering both boundary and interior cable configurations. For the inner cable arrangements, analyses were conducted on both the internal and external portions of the slab relative to the cable location. To ensure conciseness, the results presented herein refer exclusively to the configuration with  = 0.06. Fig. 3 illustrates the outcomes of the parametric analysis for Case T1, which considers a cable placed at the deck boundary. The figures capture the influence of shear connector stiffness - modeled as either rigid or flexible - on the structural response across various longitudinal deck positions defined by the dimensionless parameter  = 0.2, 0.4,