PSI - Issue 78

Marco Martino Rosso et al. / Procedia Structural Integrity 78 (2026) 301–308

303

located respectively at 2.90 m, 5.80 m, and 8.84 m high, with a net inter-story height of 2.50 m. The concrete grade used is C30/37, and steel reinforcing bars are B450C. The columns present a square cross-section of 0.20 m x 0.20 m. They were reinforced with 4 longitudinal bars with a 16 mm diameter (geometric reinforcing percentage of about 2%). Concerning their shear reinforcement, closed stirrups with a diameter of 8 mm were equally spaced every 100 mm along the entire height of the column, except for the potential plastic hinge formation area, i.e., for about 400 mm in proximity to the first floor, where they were tightened every 50 mm. Horizontal slabs are a solid RC slab typology with a total thickness of 0.40 m, with symmetrical smeared bending reinforcement bars of 16 mm in diameter. These two RC frames have already been tested in the past for an active earthquake control system device by Rebecchi et al (2023). Therefore, to make a fair comparison of the structural performances against earthquake shaking, the last floor of one of the two buildings was increased in mass with a further 14 cm of RC slab in order to account for the device weight acting on the other building. Nowadays, since that device has been removed, the last floor was increased also in the other building to provide similar mass conditions among the specimens. Moreover, internal partitions and infills were added on the last two floors. In addition, it is noteworthy that in the Building 2, a retrofitting system has been installed. Nevertheless, since it is currently under the patenting process, the authors cannot provide any further details about it. The evidence of the structure with the installed retrofitting system should be reflected by the modal properties, since this structure is stiffer than the other un-retrofitted one. These two structures are anchored to the same RC plate foundation base, so that both could be subjected to the same shaking history event. The two buildings were not structurally connected, as they were intended to be studied independently of each other. A total of 40 sensors were installed and connected to Ethernet switches. Each sensing unit is based on ASDEA MonStr devices, see Aceto et al. (2022), i.e., being a MEMS-technology-based unit equipped with a triaxial accelerometer, a gyroscope, an inclinometer, a magnetometer, and a thermometer. The power consumption of this device is extremely low (about 0.6W), and it is certified IP68 with full water and dust protection, able to operate nominally in the temperature range of -20 to 70 °C. Regarding the mounted 40 sensors, two were mounted to measure the input from the shaking table (one attached to the table directly, one attached to the RC plate foundation), and 38 were distributed evenly between the two specimens, with 19 on each. Thus, as depicted in Figure 1, the dynamic identification analysis considered only those sensors directly attached at the RC frame nodes, i.e., giving six per building and neglecting the others attached to the internal partitions and infills.

Fig. 1. Overview of the two tested RC frame buildings, and illustration of the sensing layout with its related sensors coordinates.

The two buildings were subjected to an excitation input protocol history with the adopted unidirectional shaking table as documented in Table 1. Indeed, as reported in the table, the RC frame specimens underwent to a sequence of scaled earthquake input alternating white noise inputs. The specific chosen earthquake time history was referred to the Irpinia 1980 earthquake, see Vaccari et al (1990). The first Irpinia input is designated “ Irpinia Ladder ” in the table since it consisted into a three subsequent application of the scaled Irpinia earthquake with peak ground