PSI - Issue 44

Gianluca Quinci et al. / Procedia Structural Integrity 44 (2023) 251–258

254

4

Gianluca Quinci et al. / Structural Integrity Procedia 00 (2022) 000–000

of 13.2836, and it is characterized by soil type B. Five return periods of 72, 224, 475, 975 and 2475 years corresponding respectively to the probability of occurrence in 50 years of 50%, 20%, 10%, 5% and 2% have been adopted. For this purpose, a PSHA has been conducted for deriving both hazard curves and UHS. The Matlab home made software “Mathazard” and “SCoRes”, specifically developed for the implementation of the proposed methodology, have been used herein. For details on ScoRes software the reader can refer to Giannini et al. (2022).

a) b) Fig. 1. (a) Target UHS for different return periods , (b) Seismic hazard curves of Amatrice (Italy).

The following parameters for the selection of the records from the database have been considered: magnitude range (4-8), fault-site distance range [0-100km], ground type B (according to the Italian Code). The spectra are compared in the range of periods [0, 3.0s] by using a hundred points. Figure 1a shows the generated UHS, whereas Figure 1b shows the hazard curves in terms of annual probability obtained considering or not the randomness of the GMPE. According to the method suggested in Giannini et al. (2022), the seismic risk, as shown in section 4, is calculated using avoiding the above randomness (hazard curve for e =0). 30 natural records for each return have been selected in order to have the mean and the mean plus standard deviation spectra matching the target UHSs. In this respect, scale factors have been constrained to a minimum value of 0.8 and a maximum value of 1.3. For more details on the selected records refer to Giannini et al. (2022). 4. Application to the case study and discussion of the results The record selection method proposed in Section 3 is herein used for the vulnerability and risk assessment of typical industrial braced frame primary steel structure equipped with NSCs (i.e. tanks and piping system). The full-scale mock-up consists of a primary steel structure supporting horizontal and vertical tanks, piping installations, and a cabinet. Furthermore, a conveyor was placed on the reinforced concrete slab without any connection to the main structure. It was conceived as a three-story steel frame with flexible diaphragms made of crossbeams hinged to the frame beams. Figure 2 shows the test structure on the shaking table. The ground-plan dimensions are 3.7 m x 3.7 m and the story height are 3.1 m, which leads to a total height of 9.3 m. The columns were fully clamped to a reinforced concrete base plate. This latter has dimensions of 4.8 m x 4.8 m and a thickness of 0.4 m and was constructed with concrete of class C30/37. The horizontal load-bearing system of the steel structure consisted of two braced frames (BF), equipped with tension-only bracings, in the direction of the seismic excitation and two braced frames in the transverse direction equipped with tension/compression bracings with a circular cross-section. The two BFs were connected through crossbeams, which are fastened to the frame beams by simple bolted connections with web stiffeners. These crossbeams served in turn as bearing supports for bearing platforms which were installed with bolted connections to place the secondary elements.