GNGTS 2022 - Atti del 40° Convegno Nazionale

GNGTS 2022 Sessione 2.2 257 state, and lower for the other limit states. In particular, it is observed how the effect of the sequences is more relevant for the X-CBFs than the portal frames, for whom the effects are even lower. For example, a comparison between sequences and single events is given in Figure 1.b. Finally, in order to have a more indicative estimate on the behaviour of X-CBF, it has been concluded that the following aspects are important: • The relevant influence of the seismic weight, considering a larger number of portal frames stabilized by the X-CBF; • The importance of the number of seismic events characterized by after-shocks with comparable intensity of the main-shock. Analysis on the X-CBF. The study has then be focused on the evaluation of the single X-CBF. In particular, the “BC0” model experimentally tested by Wakabayashi in (Wakabayashi et al. , 2010) has been considered (the geometrical characteristics of the connecting plates have been taken from (Faggiano et al. , 2018)). The numerical model has been implemented with the Seismostruct code (Seismosoft, 2020). First, the model has been properly calibrated on the basis of the experimental results. Then, two masses of 70 tons have been applied on the top of the columns. These masses correspond to the ones applied to an X-CBFs belonging to the same structure assessed in Section 2, but with 13 moment resisting portal frames instead of 5. In addition to the calibrated model (with H100x50x4x6 brace), two more solutions have been investigated by varying the brace profile. In total, the three following profiles have been considered: • RHS 120x40x4 ( λ – = 1,47) • H100x50x4x6 ( λ – = 2,12) • R50x20 ( λ – = 4.21) Initially, 30 natural seismic sequences have been considered. The results of the analysis show a high increase of the ductility demand for those natural sequences which have after- shocks with a comparable intensity to themain-shock ones. For this reason, artificial sequences have also be taken into account, in addition to the natural ones, with the aim to estimate the maximum possible effects of seismic sequences and to be on the safety side. For artificial sequences, two different linear combinations have been formulated, by repeating three times a single scaled event: • combination 1 (30 sequences): single event x1 – x0.9 – x0.9 • combination 2 (30 sequences): single event x1 – x0.9 – x1 The effects of natural and artificial sequences have been evaluated, by comparing the displacement requests, obtainable with the main-shock and with the entire seismic sequences. For example, shear-displacement cycles for the natural sequence of Friuli – Forgaria Cornino and his main shock are given in Figure 2, where it is clearly observable the increment of the maximum displacement demand. In order to give an overall mean assessment of the phenomenon, three different average indicators are evaluated. Different analysis group have been distinguished according to the sequence typology (natural or artificial) and the normalized slenderness (1.47, 2.12 or 4.21). The indicators are the following: • (1) in which: