PSI - Issue 44

Paola Sorrentino et al. / Procedia Structural Integrity 44 (2023) 1300–1307 Paola Sorrentino et al. / Structural Integrity Procedia 00 (2022) 000–000

1301

2

1. Introduction and motivation of the study

Through more than thirty years of practice, clear changes have occurred in the design of seismically isolated structures, accompanied by a great improvement in technologies (De Luca et al., 2022). In this regard, in Design Recommendations for Seismically Isolated Buildings by Architectural Institute of Japan (2016) the main changes occurred in the history of isolated structures in Japan have been described, identifying four different periods according to the improvements occurred. Also the authors (De Luca and Guidi, 2019) (De Luca and Guidi 2020), have given a modern key to interpret the evolution of base isolation systems, proposing a different classification of BIS design into three successive generations, as argued below. The excursion results in a modern design approach, underlining the need to cover even larger displacements ( δ ), requiring satisfactory design strategies to comply this demand. As proposed by authors for rubber solution, this can be reached by the reduction of the number of bearing points using wider devices. This means to opt for long span structural grid with few bearing points to ensure a large tributary area, or to use a transferring system which allows to reduce the number of devices in comparison to the superstructure grid. Technical improvements in this field has been encouraged also by the growing awareness in high strong motion spectral values (in terms of displacements, velocity and acceleration) recorded in real earthquakes, since 1990’s. Northridge earthquake (1994) firstly resulted in an unexpected spectral displacement of 60.48 cm at T = 2.58 s. As argued by Miyazaki (2008) only five years later, during Chi Chi Taiwan earthquake (1999) a peak ground displacement (PGD) of 289 cm was recorded. Later, Christchurch earthquake (2011) confirmed this attitude to larger displacements, reaching a spectral displacement greater than 90 cm, (Miyazaki, 2008). On the other hand, these extreme design demands have encouraged new and effective research programs, as in the case of NGA and NGA2. In particular, the “Next Generation of Ground-Motion Attenuation Models” for the western United States (NGA West) was a multidisciplinary research program coordinated by PEER in partnership with the U.S. Geological Survey and the Southern California Earthquake Center. Concluded in 2008, the NGA wanted to develop new ground-motion prediction equations (attenuation relationships) (Power at al., 2008) (Campbell et al., 2009). Resulting in new five sets of ground-motion attenuation models (Campbell-Bozorgnia, 2007) (Chiou, Youngs, 2008), the program contributed also to the creation of New NGA-East Ground Motion Database. Containing over 29,000 records from 81 earthquake events and 1379 recording stations, it is the largest database of processed recorded ground motions in Stable Continental Regions (SRCs). Later, two other NGA programs began in 2010, NGA East and NGA West 2. Resulting in a major improvement in the prediction equations, the NGA West models provide an undeniable contribution in estimating of ground motion, if compared to previous generation of attenuation models developed in 1990s and 2000s. Also in view of recent design demand for seismically isolated structures, derived from the interpretation of spectral values form recent destructive events, in this paper different sets of strong motion records have been analyzed, looking at the corresponding acceleration and displacement spectra. A comparison between results will support the evaluation In this Section, the analysed set are described and discussed. In particular, for each of them, the selected strong motion signals are indicated; their definition criteria and the highest values in terms of spectral accelerations and displacements are underlined. 2.1. Trevor Kelly (2001) In Kelly (2001) the signals are divided into two categories, pre and post 1971 strong motions. The first category refers to strong motion data from the 1940 El Centro and 1952 Kern County earthquake, El Centro and Taft Lincoln School stations respectively. Their spectra have the typical form of the earlier earthquake: the acceleration response spectrum is characterised by an amplification range up to the period of 1s, while the displacement spectrum increases with increasing period. Furthermore, their peak ground acceleration values are lower than 0.5g. Figure 1 displays the acceleration and displacement spectra of the considered stations in the two directions and their envelope. They show of design indications for BIS. 2. Set of strong motion signals