PSI - Issue 78

Annarita Palmiotta et al. / Procedia Structural Integrity 78 (2026) 489–496

493

a

b

Fig. 7. Comparisons between experimental and FEM model results about dissipated energy

Fig. 6 depicts a comparison between the experimental and numerical results in terms of horizontal force displacement imposed for all cycles. As is easy to note, the numerical results align well with the experimental data. The model successfully simulated full cycles, demonstrating a very stable force across all applied displacements. This is also confirmed in Fig. 7a, with the comparison of the energy dissipated cycle-by-cycle, calculating the area enclosed by the USDD force-displacement curve. In Fig. 6b, the percentage error between the area under the experimental and numerical curves is calculated. In particular, Error = (E num - E exp ) / E exp [%], for each cycle. The greater discrepancy in dissipated energy in the last cycle (a 40% error) likely stems from steel hardening during the USF process. This hardening causes the numerical model to predict slightly "fuller" cycles than those observed experimentally, as evident in the comparison of cycles. 4. Numerical investigations Starting from the validated FEM model, in Coniglio et al. (2025), numerical investigations are conducted in order to understand how the thickness of the connecting plates, the rib stiffness and the constraints condition could be influenced the USDDs behavior. In particular, 4 different models are implemented. The first and the second (Fig. 8a) differing each other for the connecting plates thickness (4 mm and 2 mm) and the stiffness ribs with a cross-section (Rib 1 : 55 mm and 27.5 mm; Rib 2 : 40 mm and 20 mm). The other two models are not taken into account the connecting plates and the ribs but only the constraints at the USFPs. The third model is pinned (Fig. 8b) while the fourth is fixed at the end of the USFPs. a b c

Fig. 8. Detail of FEM models for numerical investigations