PSI - Issue 44

Christian Salvatori et al. / Procedia Structural Integrity 44 (2023) 520–527 Christian Salvatori et al./ Structural Integrity Procedia 00 (2022) 000–000

521

2

1. Introduction Equivalent-frame modeling approaches are widely accepted for the global analysis of masonry buildings. Unlike more refined methods, this modeling technique allows a reasonable compromise between accuracy of results and computational effort. However, a common assumption whenever adopting this modeling technique is not to account for the out-of-plane response of masonry walls, with the underlying assumption that it is effectively inhibited by construction details or retrofit interventions. However, this assumption may be questioned in case of flexible and/or poorly connected diaphragms. For this reason, three modeling strategies were considered and compared to the experimental results of a unidirectional, incremental dynamic shake-table test on a three-story, half-scale natural stone masonry building aggregate (Guerrini et al., 2019; Senaldi et al., 2020). The first option involved a complete three-dimensional model of the experimental prototype, with an unconventional strategy to explicitly account for the out-of-plane stiffness and strength of the walls orthogonal to the shaking direction (Kallioras et al., 2019). The second one, following the common practice, ignored this out-of-plane behavior. Finally, the third solution aimed at simulating an infinitely flexible diaphragm condition, resulting in single-wall 2D models of the façades parallel to the shaking direction. Experimental and numerical outcomes were compared in terms of backbone and pushover curves, respectively, to quantify the influence of the out-of-plane behavior on the global response and to further validate the single-wall modeling approach in case of flexible diaphragms, suggested by national and international building codes (ASCE 2017; MIT 2018). 2. Benchmark experimental campaign The adopted modeling strategies were evaluated with reference to a half-scale masonry building specimen tested on the shake table at the EUCENTRE Laboratories in Pavia, Italy, within a comprehensive research project aiming at the seismic vulnerability assessment of the historical center of Basel, CH, (Guerrini et al., 2019; Senaldi et al., 2020). The specimen consisted of an aggregate of two adjacent three-story, weakly connected structural units, characterized by different roof heights and sharing a transverse party wall (Fig. 1). The dimension of the EUCENTRE shake-table imposed constructing the prototype at half scale ( λ = 0.5). To obtain physically sound results, the similitude relationship discussed by Senaldi et al. (2020) was adopted, where material densities and accelerations were not to be affected. For this purpose, the same λ = 0.5 factor was applied to lengths and stresses, whereas a λ 1/2 = 0.707 coefficient had to be used to scale time and period parameters. 2.1. Geometry and details The specimen consisted of five walls: the East and West ones oriented along the shaking direction, and the North, Centre, and South ones arranged transversely. All the façades presented several openings, except for the South and Central ones, which were completely solid. The entire prototype was 5.79-m-large and 5.58-m-wide, with the roof ridges at about 6.65 m and 7.60 m above the foundation level for the North and South units, respectively (Fig. 1).

Fig. 1. Dimensions of the half-scale masonry building aggregate specimen.