Taylor and Francis Group, European Journal of Environmental and Civil Engineering, 4(20), p. 396-411
DOI: 10.1080/19648189.2015.1039660
Full text: Download
Recent earthquakes showed that the vulnerability of some RC buildings with masonry infill walls is high when subjected to seismic actions. During the design practice, the infills are commonly treated as non-structural elements, despite their recognised contribution to the structural response of the buildings. The safety assessment of this type of buildings is important and any detailed information is essential, since such information is needed for estimation of the losses that future earthquakes are likely to induce. The present study focuses on the observation and analysis of architectural and structural design drawings of 80 buildings that represent common RC buildings with masonry infill walls located in Portugal. The data were analysed with the purpose of estimating geometric properties of the masonry infill panels confined by RC frames. Fourteen-hundred masonry infill walls were observed and were divided in to different types according to the different dispositions of the openings. Several measurements were performed in order to have deeper knowledge about the percentage of openings that exist in masonry infill walls and consequently in the building’s façade. Infill panels’ dimensions were also determined and other parameters related to the general characteristics of the buildings and the structural elements were also analysed. Parameters such as beams’ and columns’ dimensions and reinforcement details or slab thickness were also analysed. The collected data are utilized to derive probabilistic distributions, whose goodness-of-fit are partially verified with a statistical test. The results from this study can be used in structural modelling development of nonlinear models for masonry infill panels or computation of fragility models. In addition, the statistical parameters such as mean values, standard deviations, probability density functions and their goodness-of-fit have also been investigated for all the entire parameters.