PSI - Issue 44

Agnese Natali et al. / Procedia Structural Integrity 44 (2023) 2326–2333 Agnese Natali, Francesco Morelli / Structural Integrity Procedia 00 (2022) 000–000 the gap displacement of the zero-length element adopted to model the behavior of the connection exceeds the value of 1.1· 0 , then it is assumed that the deformation of the connection is so large that the upright frame loses its global stability. Fig. 3 summarizes the results of the multi-stripe analysis for the case study, using a pie chart for each intensity level. The pie charts show the number of records that experienced a failure in each stripe of analysis; if multiple failure modes are observed for a single record, the one with the highest utilization factor is considered. At the lowest intensity (i.e., 60% in 50 years hazard), the structure demonstrates excellent performance, as no member/connection failures are observed in any of the 30 records. At the 30% in 50 years level, only one record resulted in upright buckling, and the rack scored a 97% of “passing” records. The same behavior is observed in the design-level stripe (i.e., at 20% in 50 years probability of exceedance), where 1 out of the 30 records had upright buckling failure, while the rest 29 did not experience any brittle structural damage. At the 10% in 50 years intensity, which is the design level for typical steel buildings, 25 out of the 30 records passed the verification checks, while 3 experienced upright buckling of the lower level that do not employ a U reinforcement (i.e., those between 3.73 m and 12.58 m from the ground) and 2 do not satisfy the maximum plastic ovalization criteria. Similar behavior is observed for the two highest intensities (i.e., 5% and 3% in 50 years hazard), where the uprights comprised the most vulnerable components of the structure. Overall, at the design level of ARSWs (i.e., 20% in 50 years) the POS design is capable of completely preventing the diagonals from failing due to buckling or brittle failure of their connection, while the anchorage system reliably withstands the seismic uplift forces. In addition, although upright buckling is still slightly prevalent at high IMs, this happens in just one record out of 30, meaning that on average the design rules are capable of preventing fragile failures in those elements. 2329 4

Fig. 3. Multi-stripe analysis for double-depth case study DD, using a set of 30 records scaled to six IM levels (60%, 30%, 20%, 10%, 5%, and 3% in 50 years probability of exceedance). If multiple failure modes were observed on a single record, the one with the highest utilization factor is considered. 3. Experimental campaign on upright-to-diagonal connections The experimental campaign aims at evaluating the real structural behavior of the upright-to-diagonal connection designed following the POS approach. The main goals of the test campaign are two: i) at local level, evaluate the efficacy of the design rules adopted to assure the same bearing resistance when the diagonal is in tension and in compression; ii) at global level, assess the real behavior of the upright-to-diagonal sub-system. To this end two different testing setups are adopted: • “UNIVERSAL MACHINE TESTS” (UTM), where the bearing resistance of the connection is determined both in tension and in compression. This test is performed in a universal machine through which it’s possible to apply both tension and compression loads. For each diagonal, 5 different configurations for monotonic tensile tests and 5 different configurations for monotonic compression tests are tested. In these configurations, the geometrical dimensions e1 and p1 change, in order to see how they affect the bearing resistance (Fig. 4a). Displacement