PSI - Issue 78

Marta Bertassi et al. / Procedia Structural Integrity 78 (2026) 1521–1528

1523

Fig. 1. Overview of case-study walls located on the ground floor ( GF ), first floor ( 1F ), second floor ( 2F ), and third floor ( 3F ) of a URM building.

2.2. OOP wall response: geometry and mechanical modelling The OOP behaviour of masonry walls was simulated by modelling a VSSW, as shown in Figs. 2a-b, using an SDOF representation following the approach proposed by Tomassetti et al . (2018). Four configurations of masonry piers were considered, varying in geometric properties: two thicknesses ( t ) and two heights ( h ). The configurations are referred to as W1 , W2 , W3 , and W4 . For each geometry, three levels of vertical compressive stress ( σ ) were analysed. The assumed properties for all considered masonry walls are summarised in Fig. 2c. The mechanical properties of the masonry, used both in the NLKA and the NLTHA, were assumed to be identical across all configurations. The OOP response was governed by a one-way bending mechanism, idealised by a plastic hinge located at mid-height of the wall ( h 1 = h 2 = 0.5 h ), representing an initially cracked masonry element. A plastic hinge setback of 5 mm was considered at the bottom of the wall. The masonry was assumed to have a compressive strength ( f m ) of 6.7 MPa and a Young's modulus of 400 times the compressive strength ( i.e. , E m =400· f m = 2680 MPa). A zero eccentricity ( e = 0) of the overburden was assumed for all walls, and a density of 1800 kg/m 3 was considered. The parameters defining the trilinear force-displacement relationship were determined following the formulation proposed by Tomassetti et al . (2018). For the dynamic OOP analyses, energy dissipation was modelled through an analytical formulation of the coefficient of restitution (Sorrentino et al ., 2008), reduced by a correction factor of 0.85 (Graziotti et al ., 2016).

Fig. 2. (a) Undeformed configuration of the VSSW; (b) deformed VSSW; (c) geometries and vertical overburden considered in the analysis.

3. Target spectrum and accelerogram selection The elastic response spectrum selected for the NLKA is characterised by a nominal design life equal to V N = 50 years, use class II ( C U = 1), and soil category C , according to the NTC 2018 and FprEN 1998 provisions. The associated spectral parameters are a g = 0.275 g, F 0 = 2.475, T C * = 0.429, S S = 1.292, S T = 1.0, and C C = 1.386. The code response spectrum was used as the target for selecting a set of seven real ground motions (GMs) for NLTHAs. These were selected using REXELweb (Sgobba et al ., 2021), integrated with the ITACA database (ITalian