IMPLICATIONS OF MODELLING SIMPLIFICATIONS ON THE SEISMIC PERFORMANCE OF CONCENTRICALLY-BRACED STEEL FRAMED BUILDINGS
A. Silva 1, J.M. Castro 2* and R. Monteiro 1
1 Scuola Universitaria Superiore IUSS Pavia, Pavia, Italy
2 Faculty of Engineering of the University of Porto, Porto, Portugal
The research presented in this paper focuses on the implications of modelling assumptions of concentrically-braced steel framed buildings on the numerical characterization of seismic performance. A total of five case-study buildings, each reflecting a different bracing configuration typology (e.g. X, chevron), were designed following the requirements of Eurocode 8 for a moderate-seismicity location in Portugal. Several modelling variants were investigated for each of the archetypes, namely: i) non-explicit simulation of the diagonal-to-frame gusset connections (versus centreline-length with fully-pinned boundary conditions); ii) non-explicit simulation of the beam-to-column shear tab connections (versus fully-pinned boundary conditions); and iii) non-explicit simulation of the gravity-carrying frames (versus a lean-column approach). The nonlinear responses of the archetypes were compared under the design seismic intensity (i.e. Ultimate Limit State or ULS). Furthermore, fragility curves associated with the collapse and demolition limit state were derived and compared. It was found that at ULS, only variant i) had an important influence on the lateral deformation patterns of the structure. However, both collapse and demolition fragility curves indicated a relevant effect of the modelling variants on the obtained performance characterizations, generally by underestimating (gravity-frame variant) or overestimating (connection variants) fragility levels in comparison to a more refined modelling approach.
concentrically-braced frames; OpenSees; seismic performance; collapse fragility