Harmonization of the predicted post-peak seismic response of URM structures in Finite Element models

The use of Finite Element (FE) modeling approaches for nonlinear seismic analysis of URM structures has been particularly encouraged at both the research and professional levels by increasing computational power and the incorporation of advanced material models into commercial software. These approaches can accurately simulate the complex material behavior, both at the macro and micro scale, and interactions among structural elements. However, depending on the modeling scale, they may require input data, generally not available in building codes, and user expertise. Indeed, if users are not aware of the implications of the material model assumptions on results, the dispersion of all response parameters is expected to be large for different FE models. This fact is true even for small-scale benchmark structures (URM piers) unless the response in terms of lateral stiffness and shear resistance is targeted to the code-based resistance domains, as demonstrated by D’Altri et al. (2021). However, that work evidenced the occurrence of shear resistance drop for different pier drifts among the considered FE models. Note that the verification of seismic safety is highly sensitive to this aspect since the assessment methods are based on the comparison between displacement demand and capacity. In this context, the work here aims to contribute to the harmonization of FE-based modeling approaches, by addressing first the post-peak pushover response of URM piers. The implications of a predicted pier drift being compatible or not between different FE models are also shown at the wall scale. This research is framed within the DPC-ReLUIS 2022-2024 project (Subtask 10.3), funded by the Italian Department of Civil Protection. Half of the participating research groups (4 out of 8) adopted different FE modeling approaches, i.e., 3 macro-models and 1 micro-model, and different software (Abaqus, DIANA, OpenSees, SAP2000). Two piers with different aspect ratios and a two-story wall, idealized from an internal wall of the Town Hall in Pizzoli (L’Aquila, Italy), were considered as benchmark structures. The predicted pier drifts were also compared with those specified in building codes for certain failure mechanisms. The adoption of consistent assumptions at the material level, for example in terms of fracture energy, among FE models led to a good agreement of drift predictions at both the pier and wall scales, though moderate differences are observed in other response parameters.