PSI - Issue 44

Samuel Barattucci et al. / Procedia Structural Integrity 44 (2023) 426–433 Barattucci et al/ Structural Integrity Procedia 00 (2022) 000–000

427

2

1. Introduction The majority of existing buildings on the Italian territory dates back to the first half of the XX century. Hence, they were not designed to sustain seismic forces or were designed to face seismic actions lower than those expected today. As a consequence, this type of structures is not usually able to sustain the seismic actions imposed by new seismic codes and seismic zonation and seismic upgrading is needed. To select a befitting seismic upgrading intervention, the seismic assessment of the existing structure must bring to light the structural deficiencies appropriately. To this end, numerical models are developed to provide an accurate prediction of the structural response, keeping affordable computational costs. Because of this, the contribution to the structural response due to non-structural elements was often ignored, because their effect was deemed negligible. However, several numerical and experimental studies have shown that masonry infills substantially influence both stiffness and strength of the frame. In particular, Dolsek and Fajfar (2008) showed that the maximum base shear coefficient of the infilled frame is significantly larger than that of the bare frame. In the case of the infilled frame, the maximum force is reached at a top displacement smaller than that corresponding to the maximum force in the bare frame and, for a given peak ground acceleration, the top displacement reached by infilled frames is smaller than the corresponding value for the bare frame. This means that the stiffness of the infilled frame is larger than that of the bare one, as also demonstrated by Manfredi et al (2012) with three dimensional numerical models. Experimental tests by Negro and Verzeletti (1996) and by Colangelo (2005) also showed that the initial stiffness of a bare frame may increase by one order of magnitude when infills are considered, and the peak strength may become more than double. Based on this evidence, an accurate prediction of the structural response needs to take into account properly the contribution due to such non-structural elements. To this end, the paper aims at presenting a technique to calibrate a finite element numerical model of r.c. frames with infill panels to match the results of a quasi-static cyclic experimental test of a prototype infilled frame. The proposed technique follows a sub-assembly approach and stems from the experimental evidence that the response of the structure, subjected to small amplitude displacements, is mainly governed by the infill panel, while it tends to become coincident with that of the bare r.c. frame under large amplitude displacements. The validity of such technique is independent of the type of elements adopted to simulate structural and nonstructural members and its use could be extended also to other numerical models available in literature for infills. The paper develops into two main parts: first, a case study one-bay one-storey r.c. infilled frame has been designed and tested. In the second step, a numerical model has been built and calibrated following the proposed approach so that the prediction of the numerical model match well with the response provided by the experimental test. The cyclic responses are compared in terms of base shear and top displacement. 2. Design of the case study frame The structural system analysed by numerical analysis and experimental test is a one-bay one-storey r.c. frame (Fig. 1a). This frame is the central span of the first storey of a four-storey r.c. building which was designed to be representative of typical Italian residential buildings of the seventies. The length and the height of the frame are equal to 4.00 and 2.95 m, respectively. According to the regulations in force during the seventies in Italy (Italian Ministry of Public Work (1971), Ministry Decree (1974)), only gravity loads are considered for the evaluation of the design internal forces of structural members. Dead loads g k and live loads q k per unit area are determined considering the nominal values provided in (Ministry Decree (1996)) The characteristic values of g k and q k of the slab are here assumed equal to 5.8 and 2.0 kN/m 2 , respectively. The weight of the beam is equal to 3.75 kN/m, which is evaluated considering a cross section having dimensions equal to 30 x 50. The internal axial forces N acting on columns and the distributed loads resting on beams are evaluated based on the tributary area concept. Columns are designed to resist axial force only, while beams are sized to sustain bending moment and shear force. Size of the cross sections and area of the steel reinforcements of columns and beams are determined by means of the allowable stress method according to the provisions of the building code in force in the seventies (Ministry Decree (1974)). The minimum reinforcement ratio prescribed for the tension zone of beams is equal to 0.0015. The minimum area A s of the