GNGTS 2021 - Atti del 39° Convegno Nazionale

GNGTS 2021 S essione 2.2 312 • glass: E g =70GPa, ν g =0.22 and γ g =2500 kg/m 3 ; • PVB: G PVB , ν PVB = 0.49 and γ PVB = 1000 kg/m 3 . To obtain an accurate structural response there is an essential need to incorporate the interaction between glass panel and frame and the gasket friction. A simple axial connector was assigned to wire feature which contains wires connecting points from LG element to transom. Actually, an infinite flexural stiffness of transoms was considered to represent the effective behaviour of this type of wall (i.e., with a rigid connection between panel and the structural background). In addition, the in-plane stiffness (K x and K y ) of each connection was calculated in function of the modulus of elasticity E adh =2.4 MPa, the mesh dimensions, the length and the thickness of the connection, and the elastic stiffness in the z-component was considered as infinite, thanks to the presence of the two steel angles in correspondence of the transoms. The mesh has been assigned to the various elements in order to reach a good compromise between the computational speed of the model and the accuracy of the results, considering that the two types of elements are connected via connectors. 3 Methodology: Cloud Analysis In the last decades, the Seismic Probabilistic Risk Assessment (SPRA) appears to be the most appropriate methodology to investigate the seismic risk which is the combination of vulnerability, exposure and hazard. The basic principles underlying of this approach are seismic hazard curves and fragility curves of a particular structural element, substructure or a whole system. Moreover, the quality of the results depends on the readability of fragility analysis method. Porter et al. (2007) in 2007 presented the seismic fragility functions which can be used to predict a realistic estimation of a certain level of damage that a glass system would tolerate due to a given earthquake event. In literature, several examples are available in function of the type of failure, the preliminary assumption and the level of detail (Zentner et al.,2016), such as Safety factor method, Numerical simulation using linear regression in log-space (Cloud Analysis), Incremental dynamic Analysis (IDA). The methodology selection is strictly dependent on the available data on site, structure and any reasonable probabilistic model to consider epistemic uncertainties (i.e., uncertainties in component capacities, in material properties and construction details). In this paper, a modified version of Cloud Analysis (Jalayer et al.,2014) considering a set of 60 unscaled ground-motion records, chosen from European Strong Motion Database, has been implemented to reduce the computational effort of a typical IDA analysis. It is based on coupling linear regression in log-space of structural response versus intensity measure of seismic event. Finally, fragility curves allow engineers to estimate the probability of attainment of a definite damage state (e.g. glass fallout) given a specifiedmagnitude of an engineering demand parameter, EDP i , (e.g. interstory drift ratio) and are presented in the form of cumulative distribution function by Eq.(1) with respect to the IM i . thodology: Cloud Analysis the last decades, the Seismic Probabilistic Risk Assessment (SPRA) appears to be the ppropriate methodology to investigate the seismic risk which is the combination of ability, exposure and hazard. The basic principles underlying of this approach are c hazard curves and fragility curves of a particular structural element, substructure or e system. Moreover, the qu lity of the re ults depends on the readability of f agility is method. Porter et al. (2007) in 2007 presented the seismic fragility functions which used to predict a realistic estimation of a c rtain level of damage that a glass system tolerate due to a given earthquake event. In l teratu e, several examples are available ction of the ype of failure, the preliminary assump ion nd the lev l of detail er et al.,2016), such as Safety factor m thod, Numerica simul tion using li ear ion in log-space (C oud Analysis), Incremental dynamic Analysis (IDA). e methodology selection is strictly dependent on the vailable data on site, structure ny reasonable probabi istic model to consider epistemic uncertainties (i.e., inties in compon nt capacities, in material properties and construction details). this paper, a modified version of Cloud Analysis (Jalayer et al.,2014) onsid ing a 60 unscaled ground-motion records, chosen from European Strong Motion Databa e, en implemented to reduce the computational effort of a typical IDA analysis. It is on coupling linear r gressio in log-space of structural esponse v rsus intensity re of seismic event. ally, fragility cu v s allow e gineers to estimate the probability of attainmen of a e damage state (e.g. glass fallo t) given a specifi d magnitude of an engineering d parameter, EDP i , (e.g. interstory drift ratio) nd are presented in the form of ative distribution funct on by Eq.(1) with respect to the IM i . F i ( EDP i ) = ¿ Φ ( ln ⁡ ( EDP i θ i ) β i ) (1) : i is the median for IM given EDP i ; i is the logarithmic standard deviation for IM given EDP i ; ubscript i represents the damage state of interest. ud Analysis S ismic Probabilistic Risk Assessmen (SPRA) appears to be the logy to investigate the seismic risk which is the ombinat on of d hazard. The basic principles underlying of thi approach are fr gility r s f a p rticular structur l element, ubstr cture o r, the quality of th results depends on the readab lity of fragility al. (2007) in 2007 presented the seismic fragility functions which alistic estimation of a certain level of amage that a glass system en e rthq ake event. In li eratur , several examples re avai able of failure, the reliminary assumption a d the lev l of detai h as Safety factor method, Numerical simulation using linear l ud Analy is), Incre ental d na ic A alysis (IDA). ction is strictly d pendent on the available data on site, structure obabilistic model to consider epistemic uncertainties ( .e., t capac ties, in material prop rties and construc ion details). ed version of Cloud Analysis (Jalayer et al.,2014) considering a -motion r cords, chosen from European Strong Motion Database, reduce the compu ation l effort of a ypic l IDA analysis. It is regressio in log-space of structural resp nse versus intensity s allow eng e s to timate the p obability of attainment of a . glass fallout) given a specified magnitud of an engineering , (e.g. interstory drift ratio) and are presented in the form of nction by Eq.(1) with respec to the IM i . P i ) = ¿ Φ ( ln ⁡ ( EDP i θ i ) β i ) (1) giv n EDP i ; tandard dev ation for IM given EDP i ; the damage stat of interest. (1) Where: • θ i is the median for IM given EDP i ; • β i is the logarithmic standard deviation for IM given EDP i ; • subscript i represents the damage state of interest.