PSI - Issue 44

Fabio Mazza et al. / Procedia Structural Integrity 44 (2023) 147–154 Fabio Mazza et al. / Structural Integrity Procedia 00 (2022) 000–000

153

7

fluctuation of values can be noticed for elevated tanks too (Figure 6b). However, in the light of the simplified modelling assumptions (i.e. elastic piping elements and equivalent elastic beam with lumped mass tanks), numerical results demonstrate an acceptable level of approximation. The structural dynamic matching proves to be fundamental to evaluate realistic maximum floor accelerations used to calculate DAFs.

(a) Piping system. (b) Elevated tank. Figure 6. Dynamic amplification factor peak values for the fixed-base building.

4.3. Base-isolated building

Nonlinear dynamic analysis of the base-isolated steel framed structure is carried out by using a self-developed computer code (Mazza and Labernarda (2020)), modified in order to handle the simplified (SM) and advanced (AM 1 , AM 2 ) models of CSSBs. The nonlinear behaviour at the end sections of frame members is considered by means of a lumped plasticity model. Shear deformation of frame members is neglected, while axial and flexural stiffness are evaluated considering a Young modulus equal to 210000 and 34313 MPa for steel and RC members respectively. Mass and stiffness proportional damping is assumed, with a viscous damping ratio equal to 1%.

(a) OS-2_50XY, X direction.

(b) OS-2_50XY, Y direction.

(c) JMA_Kobe_50XY_Z100, X direction. (d) JMA_Kobe_50XY_Z100, Y direction. Figure 7. Displacements time-histories along X and Y directions for the base-isolated building, considering SM, AM 1 and AM 2 for CSSBs.

The displacement of the isolation level along X (d X ) and Y (d Y ) directions, calculated by averaging displacements of the monitored nodes 1 and 4 shown in Figure 2, is evaluated and reported in Figure 7, assuming three models of CSSB (i.e. SM, AM 1 and AM 2 ), together with the maximum and minimum experimental recorded values. Notable differences are highlightable with reference to the medium intensity (50%) OS-2_50XY earthquake (Figures 7a,b), especially along the X direction. In particular, SM (blue curve) results are significantly different from those obtained using AM 1 (red curve) and AM 2 (green curve) and underestimate the displacement of CSSBs along X and Y directions. Instants of time and values of maximum and minimum experimental response are well predicted along the X direction, with a slightly worse prediction along the Y direction. In the case of the medium seismic intensity (50% along X and Y directions and 100% along the vertical one) JMA_Kobe_50XY_Z100 earthquake (Figure 7c,d), not so important differences in terms of displacement are evident among the three considered CSSBs models during the initial part of