PSI - Issue 78

Tahir Ahmad et al. / Procedia Structural Integrity 78 (2026) 631–638

633

in Fig. 1b. The steel material properties used in the model are provided in Table 1. To ensure high mesh quality and computational efficiency, the components were divided into multiple parts before meshing. Solid elements of type C3D8R (8-node linear brick elements with reduced integration and hourglass control) were used, with a refined mesh size of approximately 5 mm to accurately capture local deformations and stress concentrations. A Dynamic Explicit analysis step was selected to effectively manage the complex contact interactions and material nonlinearities characteristic of these connections. After successful validation, the calibrated FE model was employed to generate additional data points for the moment – rotation ( M – θ ) curves. Fig. 2a presents several experimental M – θ curves. Fig. 2b shows the bilinear idealization of the M – θ response, developed using the equal energy method as recommended by EN 15512 (2020). The design rotation was selected to ensure that the area under the bilinear curve matched that of the original curve within ±5%, thereby maintaining equivalent energy dissipation capacity. To account for the interaction between moment and axial force, a linear interaction domain was assumed, as illustrated in Fig. 2c. 2.2. Column base-plate connection subjected to biaxial bending and high compression axial forces Following the successful validation of the beam-to-column connection models, detailed three-dimensional finite element models were developed in Abaqus (2024) to accurately capture the complex axial force – biaxial bending ( PMM ) interaction and moment – rotation behavior of the upright column connected to the base. These models incorporated the intricate effects of bolt connections and local buckling in the perforated uprights. The element types and meshing strategy were consistent with those previously established and validated for the beam-column connection simulations. Loads were applied at the top of the upright, and reactions were collected at the base using distinct reference points (RPs). These RPs were kinematically coupled to the top and bottom surfaces of the upright via rigid body constraints, ensuring consistent and realistic load transfer. Fig. 3 presents the Von Mises equivalent stress contours for the base-plate connection subjected to axial force ( P ) and bending about either the X-direction (Fig. 3a) or Y-direction (Fig. 3b). In both loading cases, the stress contours revealed significant concentrations and localized yielding, particularly in the thin-walled regions of the upright around perforations and at the base-plate interface. These observations highlight the primary failure zones governed by local buckling and material yielding. The corresponding moment – rotation ( M – θ ) curves are shown in Fig. 3c. The response in the X-direction reaches a peak moment ( M y ) of approximately 9.5 kN·m, followed by a more gradual and ductile softening. In contrast, the M x curve (bending in Y-direction) shows a higher peak moment of about 13 kN·m, but exhibits lower rotational ductility and a sharper drop in capacity. This directional disparity reflects the influence of the upright's non-symmetrical cross-section and the specific configuration of the base-plate connection on its stiffness and resistance.

Table 1. Material properties of steel specimens. Mechanical properties

Value

Elastic modulus (MPa)

210000

Poisson Ratio

0.3

Yield Strength (MPa) Ultimate Strength (MPa)

350 420

(a)

(b)

Fig. 1. Beam-column connection: (a) Details of the specimen tested; (b) Von Mises stresses map.