PSI - Issue 78

Giuseppina Uva et al. / Procedia Structural Integrity 78 (2026) 1048–1055

1052

provided by the experimental campaign or that were very dispersed). Finally, the comparison of the 1 st vibration mode (that was the one identified by field dynamic tests, Pelà et al., 2013) shows a nearly perfect correspondence. In the Shell Box Model, the geometry is defined by fictitious shell elements, and in order to guarantee the “equivalence” with the full model in terms of the modal response the thicknesses assigned to the different plane elements is properly calibrates after an iterative process in which the thicknesses are varied, modal analysis is performed and comparing modal parameters compared until the best fitting is reached.

Figure 3. Materials assigned to the different elements of the Solid Model – SM (a) and of the Shell-Box Model (b).

The comparison has been performed with regard to the first three vibration modes. The 1 st mode, or fundamental mode, is usually the most interesting one, since the largest displacements caused by seismic actions are mainly out of plane and consistent with the 1 st modal shape. For it, the difference observed in terms of frequency and period is within the +/- 10% range. In the next modes, instead, the differences on period are larger. However, this does not significantly affect the determination of the spectral pseudo-acceleration, which, in fact, exhibits a negligible variation on the horizontal and vertical components for all three modes. The shape of the first three modes is the same for both models (Figure ), demonstrating that the reduced SBM retains the information about the dynamic behaviour of the structure and can be an effective alternative for performing both dynamic identification procedures and cost-effective non-linear dynamic analyses.