Assessment of existing steel frames: Numerical study, pseudo-dynamic testing and influence of masonry infills

摘要

Most of existing steel multi-storey frames in Europe have been designed before the introduction of modern seismic design provisions, hence they often exhibit low performance under earthquake loads due to their low lateral resistance and energy dissipation capacity. In addition, such structures often include rigid and brittle masonry infill walls that highly influence their lateral response and distribution of damage pattern. However, current procedures for the assessment of existing steel buildings in Europe, included in the Eurocode 8 - Part 3 (EC8-3), do not provide adequate guidance for the assessment of 'weak' steel frame with masonry infill walls. Moreover, most of available modelling approaches of masonry infills formerly developed for reinforced concrete (RC) structures do not properly represent the behaviour of infill walls in steel frames. An improved numerical has to be provided to satisfactorily mimic infill walls' behaviour in steel moment frames. To this end, an experi-mental and theoretical study was carried out within the framework of HITFRAMES (i.e., HybrId Testing of an Existing Steel Frame with Infills under Multiple EarthquakeS) SERA project. This paper firstly presents the limitations of current EC8-3 by conducting a code-based assessment on a case study steel moment frame using pushover analysis. Three different single strut models, widely used for simulating the presence of masonry infills in RC structures, are considered for the numerical analyses. The paper also presents the results of pseudo-dynamic (PsD) tests performed on a large-scale 3D steel frame with masonry infills. The capability of the different masonry infill models is successively evaluated by comparisons between numerical and experimental results. On the basis of the obtained results, recommendations on how to potentially improve the single strut model for masonry infills surrounded by steel frames are also provided.