PSI - Issue 44

Sergio Ruggieri et al. / Procedia Structural Integrity 44 (2023) 1964–1971 Sergio Ruggieri et al./ Structural Integrity Procedia 00 (2022) 000–000

1967

4

buildings according to the seismic hazard level of the area under investigation, it is necessary to know the localization of the building. Gravity loads (G and Q) are also unknown, but they can be defined according to the destination of use of the building under study.The simulated design process can be performed by referring to the code prescriptions of the year of construction, which is an unknown parameter in the proposed procedure. Depending on whether seismic details are present or not, beams and columns can be designed taking into account seismic actions or considering only gravity loads. Considering that we are dealing with existing buildings and that the geographical context is Italy, up to 1980 the presence of any seismic detail should be excluded; from 1980 to 2008 the presence of poor seismic detail can be considered, after 2008 the presence of full seismic detail can be considered. In the first two cases, the admissible stress method should be considered as the design method, while in the last case, design can be performed through limit states approach. The description of simulated design procedures for buildings with complete seismic details (e.g., the capacity design approach) is skipped, as it is widely known. It is worth mentioning, instead, the procedure for the simulated design of primary structural elements with the allowable stress method. By assuming a fixed value of the homogenization coefficient (10 to 15), columns are designed with a simple axial stress and beams are designed by defining the maximum bending stress in the main sections. The evaluation of the structural elements size is performed by automatically subdividing the sides of the building in equal parts and imposing a maximum value for the bay lengths equal to 5 meters. Interstorey height can be also varied into the models, especially when the presence of a higher ground floor is detected by Bi VULMA . In general, if the building is residential, an interstorey height of 3 ÷ 3.5 m can be assumed. Once structural element details are defined, a set of mechanical models can be derived by combining all parameters. The number of models grows as the discretization of the parameters increases, as shown in Leggieri et al. (2021). For the present case, the models of RC buildings can be generated and analysed automatically using open-source software, such as Opensees (McKenna, 2011). 3D models are used, whereas, considering the uncertainty related to the input parameters, simple modelling approaches are adopted for accounting for nonlinearities. Beams and columns are modelled as one-dimensional frame elements with fixed restraints at the base. A rigid diaphragm is positioned on each floor. Nonlinearities are simulated through a lumped plasticity approach by placing plastic hinges at the end sections of structural elements and ensuring cyclic degradation for nonlinear dynamic analysis. Inelastic ductile mechanisms are assumed, accounting for the combination of axial and bending stresses for columns (bidirectional hinges) and only bending for beams. Eurocode 8 prescriptions are followed for hinge constitutive laws, while shear failures can be checked a posteriori for each structural element. Masonry infills are modelled using two diagonal struts in each frame, where nonlinearities are defined according to the classical formulations proposed in Panagiotakos and Fardis (1996) and considering only in-plane mechanisms.

Fig. 1. Framework of the proposed procedure.