PSI - Issue 64

1972 E. Michelini et al. / Procedia Structural Integrity 64 (2024) 1967–1974 6 Elena Michelini, S ł awomir Dudziak, Simone Ravasini, Beatrice Belletti / Structural Integrity Procedia 00 (2019) 000–000

Seismostruct curve. As can be seen, all limit states are associated to the reaching of the corresponding steel strains, whereas threshold values of concrete strains are not achieved (not even that corresponding to LSC2). This is related to the low reinforcement ratio adopted in those parts of the structure that are subjected to tension in the accidental situation (due to the change of the bending moment side). Consequently, the concrete capacity cannot be fully utilized. Other results of numerical analyses are summarized in Figure 6, with reference to the model without roof developed in Seismostruct. The evolution of the vertical reactions and monitored angles with increasing settlements are shown in Figure 6(a). The force redistribution from the central column to the side ones can be observed. Deformed plots with indication of damage concentration at different LS are shown in Figure 6(b). For the scenario of middle column settlement, besides the development of plastic hinges near the frame’s nodes, loss of global stability also occurs, as indicated by the sudden increase of the total inclination α .

(a) (b) Fig. 5. Comparison among different models in the case of the middle column’s settlement (scenario No.1), in terms of: (a) sum of reactions vs. settlement curve, (b) diagram of normal forces for a settlement value u = 90 mm.

(a) (b) Fig. 6. Settlement of the middle column: (a) reactions or selected angles vs. settlement, (b) deformation (10x scaled) at the considered limit states with damage concentration indicated. The same procedure discussed above is repeated for the settlement scenario No.2. The results of preliminary analyses are shown in Figure 7. Also in this case, for the model with roof beams, an alternative load-bearing scheme develops, in which roof-beams act like a suspension system for the column subjected to large settlement. For the same reasons discussed before, the model without roof beams is selected to perform the detailed analysis. For the latter model without roof, Figure 7(a) provides a further comparison between the numerical results obtained from Seismostruct and ABAQUS software, in terms of sum of reactions under the fixed columns vs. settlement of the side column. In this case, the curve obtained with ABAQUS slightly underestimates the initial stiffness due to fact that the tensile strength of concrete is neglected. Once again, all the LS are reached due to the achievement of limit values of steel strains, whereas threshold values of concrete strains are not achieved. Other results of numerical analyses are summarized in Figure 8 and Table 2, with reference to the model without roof in Seismostruct. From Figure 8(a), it can be inferred that a force redistribution takes place from both side columns to the central one. From Figure 8(b),