PSI - Issue 78

Paolo Ielpo et al. / Procedia Structural Integrity 78 (2026) 1024–1031

1028

Fig. 3. illustration of load application steps in the 3D numerical model.

The final step of the modeling process involved defining the mesh: a structured grid composed of eight-node hexahedral elements was adopted for the concrete, while a linear mesh capable of supporting only axial loads was used for the reinforcement. The overall model comprises 4,441 nodes, divided into 173 linear elements, 2,645 hexahedral elements, and 1,710 tetrahedral elements, the latter employed for the steel plates. The average size of the hexahedral elements, set at 0.06 m, was determined through a sensitivity analysis. 4.1. Calibration procedure of the numerical model To develop an accurately calibrated numerical model, a detailed analysis was conducted on the cyclic test results of the T4 specimen. The analysis focused particularly on the failure mechanisms and the load-drift behavior observed during the cyclic test. The experimental load-drift curve of the T4 specimen showed a peak load capacity of 42.9 kN at a drift value of 0.93%. The ultimate drift, defined as the point at which the load decreases by 20% from the peak value, was measured at 3.45%. Between 1% and 3% drift, the specimen exhibited stable post-peak behavior, while beyond 3% drift, a rapid degradation of load-bearing capacity occurred. This deterioration was attributed to instability phenomena in the bottom longitudinal reinforcement beam bars (Fig. 4).

a)

b)

Fig. 4. cyclic loading test: a) failure mechanism b) experimental results