GNGTS 2023 - Atti del 41° Convegno Nazionale

Session 2.2 GNGTS 2023 Comparison between numerical and empirical derivation methods of component fragility functions S. Mattei, C. Bedon University of Trieste, Department of Engineering and Architecture, Trieste, Italy Introduction During the last years, there is growing evidence to suggest that the performance of non-structural elements affects the whole structural behaviour. Thus, the investigation of seismic behaviour and the definition of the probability of the structural damage are essential to reduce the risks involved. In the construction field, the glass structures have been chosen for their innovative characteristics: lightness, brightness, harmony and adaptability in both natural and urban contexts (Wurm, 2007) although the brittle behaviour of glazing systems under any type of load (quasi-static, dynamic, etc.) has always led to financial losses or injuries and deaths. As a consequence, the structural response of such systems under accidental actions has been investigated in many previous works (Mattei S. and Bedon C., 2021; Mattei et al. , 2021). The identification of the mechanical performance supports in assessing probabilistic seismic risk and classifying the buildings (Jalayer et al. , 2011). In this context, fragility curves are widely used in estimating a certain level of damage for a specific structure, element or sub-structure. In the present study, a comparison between numerical and empirical derivation methods of component fragility functions is presented, in order to assess the reliability of results that can be obtained via the so-called Cloud procedure. In particular, the obtained cumulative probability functions are determined with the aim to improve the computational efficiency by combining different geometries and applications. Case-studies This section presents an implementation of the Cloud methodology in the case of different curtain wall configurations by varying glass-to-frame clearance from 3 mm to 11 mm. The chosen