PSI - Issue 84

Paolo Andrea Miglietta et al. / Procedia Structural Integrity 84 (2026) 1111–1118 P.A. Maglietta et al. / Structural Integrity Procedia 00 (2026) 000–000

1117

7

4.1. Numerical analysis of the pier

Aiming to assess the effect of degradation phenomena on the seismic behaviour of the case study pier, the obtained combinations of M- χ and V - γ were implemented in a numerical non -linear model. Hence, non-linear Push-over (PO) analyses were carried out at different time periods starting from construction time. A bidimensional (2-D) numerical model of the case study pier was developed using the OpenSees platform (McKenna et al., 2000). The pier shaft was discretized into four elements, each represented by nonlinear beam-column elements. A refined discretization was adopted at the base of the pier by increasing the number of integration points, in order to simulate more accurately the nonlinear curvature along the pier axis. A fixed support was assigned to the bottom section by constraining the displacements (U x , U y ) and rotations (R z ). The cross-sectional response at each integration point was defined by pairing the M- χ and V - γ relation ships previously obtained, while elastic axial behaviour was assumed. The vertical load on the pier was applied on the top node and was kept constant. The PO analysis was carried out in displacement control, by setting the top node as control node. The target displacement was set equal to 350 mm. Fig. 4a-c shows the PO curves associated with the considered percentiles. The curves associated with P16 (Fig. 4a) exhibit significant variability in both load-bearing capacity and ultimate displacement. Additionally, starting from approximately the 90 th year, a change in the failure mode can be observed: structural collapse is triggered by shear failure rather than by flexural failure. Conversely, the curves corresponding to the P50 and P84 (Fig. 4b-c) are nearly identical and do not show any change in the failure mechanism. Such results highlight that the structural behaviour of RC members is highly influenced by the randomness of the input parameters considered. Lastly, the seismic capacity expressed in terms of dissipated energy over time for each considered percentile is illustrated in Fig. 5. The latter was computed as the area under the PO curve up to the peak strength. The capacity of the damaged pier, evaluated using the M- χ and V - γ relationships corresponding to the P16, exhibits a significant drop over time. In contrast, nearly constant values were obtained considering both P50 and P84. These results further emphasize that the uncertainties associated with environmental conditions and building characteristics play a key role in the structural assessment of damaged components.

Fig. 4. Time-dependent capacity curves of the pier: (a) 16 th percentile (P16); (b) 50 th percentile (P50); (c) 84 th percentile (P84).

Fig. 5. Seismic capacity expressed in terms of energy associated with the three percentiles considered.