The behaviour and safety of precast systems predominantly depends on the performance of the connections between the precast elements. During the recent earthquakes in the Emilia-Romagna region in Italy, the cladding-to-structure connections, which were installed in industrial precast buildings, did not perform as it had been expected in their design. Several cases of collapse of cladding panels was observed. This indicated the need for new research and improvements in the design and execution of such connections. In this study the cyclic response of the cladding-to-structure connections, which are most commonly used in the current European design practice of precast buildings, was experimentally and analytically investigated. Such connections consist of cold-formed or hot-rolled channels and hammer-head straps. They provide the out-of-plane restraint of vertical or horizontal panels. Until now no information has been available, in the existing literature, about the cyclic response of such connections in the plane of the cladding panels. Therefore extensive relevant experimental studies were performed. Results were used to develop reliable numerical models of typical connections, and to define appropriate procedures for their design. This knowledge was used to conceive the shaking table experiments in Skopje.
COBISS.SI-ID: 7456865
A second-line back-up system for the seismic protection of cladding panels in RC precast buildings is presented. The system consists of special anchoring elements and a rope restrainer. The latter is activated only in the case when the existing connections between the primary structure and the panel fail, resulting in the occurrence of large impact forces in the restrainer and in the anchoring elements. In order to adequately design the constitutive parts of the system, a simple yet sufficiently accurate procedure for the estimation of the impact forces is needed. A relatively easy-to-use formula was therefore proposed for this purpose. Next, an extensive parametric study, using response history analysis (RHA), was performed and the influence of several parameters affecting the impact forces in the restrainers was studied. The results obtained in the study were used to evaluate the proposed analytical formula. Considering the simplicity of the proposed formula, its accuracy was good.
COBISS.SI-ID: 7581793
The results of the shaking table experiments, which are described in this report, were published for the first time.
COBISS.SI-ID: 8490081
Based on the experimental and analytical studies of complex response of typical cladding panel connections, appropriate robust engineering numerical models were defined, which can be used in the design practice. The design procedures were proposed, and the expressions, which can be used to estimate their displacement and strength capacity, were derived.
COBISS.SI-ID: 7949153
Fragility curves for six classes of precast buildings are presented. Only the type of non-structural components was varied in order to define different variants of investigated building classes. Precast buildings without non-structural components represented a base case of investigation. In addition, one building class with vertical panels, one with masonry infills and three building classes with horizontal panels were defined, each with a different type of panel-to-structure connections. Each building class was represented by a sample of 100 buildings. For each building a simplified three dimensional numerical model was built, in order to perform a large number of non-linear dynamic analyses, but still capable to account for phenomena such as panel-to-structure interaction, infill-to-structure interaction, sliding of the unrestrained elements and accounting for the non-structural components only until the point of their dislocation. Based on the simulations, fragility curves for three non-structural damage states and two structural damage states were developed. It can be observed that the non-structural components had a detrimental impact on all levels of damage, including structural collapse. In addition, the effect of non-structural components is illustrated using an example of a non-linear dynamic analysis. It is shown that the mass of non-structural components had in most investigated cases detrimental effect on damage, whereas the effect of increased stiffness proved to be more unfavourable if the non-structural components are attached to the structure below the beam-to-column connection rather than above it.
COBISS.SI-ID: 7994209