PSI - Issue 78

Ahmed Mabrouk et al. / Procedia Structural Integrity 78 (2026) 960–967

962

using Steel02, which includes kinematic hardening and Bauschinger effects to simulate cyclic degradation accurately. A Gauss – Lobatto integration scheme with five integration points per element is used to balance computational efficiency with fidelity in capturing nonlinear effects.

Fig.1. Donà et al, (2021) Macro-model of infill panel: a) general presentation and b) detail of the BWH element

3. Infill macro-model calibration The calibration process is based on experimental data from two benchmark studies that investigated the coupled IP and OOP behavior of infilled RC frames. The first, by Calvi and Bolognini (2001), tested unreinforced masonry (URM) panels under cyclic IP loading followed by monotonic OOP loading, capturing the significant reduction in OOP capacity caused by prior IP damage. The second, by Minotto et al. (2021), examined retrofitted panels using three different TRM/FRM systems — F, FB, and RBB — under a similar testing protocol. In each case, OOP tests were conducted after three levels of IP drift: 0%, 0.5%, and 1.2%. While the macro-model was calibrated against all panel types (URM, F, FB, RBB), only the URM and F configurations are presented here, as they are directly relevant to the dynamic analyses performed in the case study. The re-implementation and recalibration of the model, based on the formulation by Donà et al. (2021), was carried out within the STKO – OpenSees framework, ensuring consistency in modeling assumptions and allowing for seamless integration into the time history simulations. The model successfully captured the distinct behaviors of the URM and F panels, including the increase in both IP and OOP capacity offered by the strengthening system. The F panel demonstrated improved resistance due to the added plaster layer, with the model accurately reproducing its enhanced performance across varying levels of prior IP damage. Based on the calibration results, a set of displacement interaction domains was constructed (see Table 1), linking the OOP capacity of each panel to its corresponding IP drift level. These domains form the basis for defining infill performance limit states, allowing the model to simulate the progressive degradation of OOP strength under seismic loading. This calibration confirms the model's ability to reproduce realistic infill behavior and supports its application in the nonlinear analyses presented in the next chapter.

Table 1. IP drift values at limit states, and OOP displacements as a function of IP drift at OOP limit states URM F IP drift (%) (DLS, , ) 0.3%, 0.5%, 1.0% 0.3%, 0.5%, 1.5% OOP disp. [mm] DLS 5.10 14.60 16.00 + 6.4 6.25 +16.00 30.00 + 8.0 7.08 +17.5