Energy‐Based Characterisation of Earthquake Damage in Masonry Infills Through Combined In‐Plane and Out‐of‐Plane Experimental Tests

The seismic capacity of structural systems/members is usually defined in terms of deformation thresholds. Although this is effective in many circumstances, disregarding cumulative quantities such as hysteretic energy may be ineffective when low‐cycle fatigue is relevant (e.g., long‐duration ground motions, very soft soils, ground‐motion sequences). This paper presents the outcomes of an experimental campaign designed to explore the suitability of hysteretic energy thresholds to characterise damage. To do so, three nominally identical masonry infill panels were subjected to in‐plane (IP), quasi‐static, cyclic displacement‐controlled loads generating a moderate level of damage, followed by out‐of‐plane (OOP), dynamic shaking‐table tests amplified until collapse. The study investigated the effect of the IP hysteretic energy dissipation on the observed IP damage state, and the subsequent OOP collapse capacity. Different IP loading protocols were designed to induce the same peak amplitude‐based engineering demand parameters (EDPs) while modulating the energy‐based demands. In the IP phase, two specimens showed a damage pattern dominated by sliding along two to three horizontal cracks, while the third specimen was dominated by a diagonal, strut‐like failure. For the sliding‐dominated specimens, the IP hysteretic energy showed signs of a positive correlation with the level of damage, and, in turn, a possible negative correlation with the peak floor acceleration leading to complete OOP collapse. On the one hand, further complementary experimental tests and/or refined numerical analyses are needed to generalise the trend highlighted here. On the other hand, the presented results pave the way for using energy‐based EDPs for state‐dependent seismic fragility assessment methodologies, thus allowing a refined consideration of damage accumulation.