PSI - Issue 44

Elena Michelini et al. / Procedia Structural Integrity 44 (2023) 1530–1537 Elena Michelini et al / Structural Integrity Procedia 00 (2022) 000 – 000

1533

4

3. Ambient vibration tests and dynamic characterization of the building The structural vibration of Borgo Val di Taro Town Hall, due to ambient noise excitation, was measured using 30 monoaxial piezoelectric ICP® accelerometers, model 393B31, produced by “ PCB piezoelectric ” . The accelerometers are characterized by a measurement range of ±0.5 g and a resolution of up to 10 -6 g, in the frequency band between 1 and 10000 Hz. The accelerations of the measured points were recorded according to two orthogonal horizontal directions X and Y by screwing two mono axial sensors onto an aluminum cube. Five “ Simcenter SCADAS XS ” dataloggers were used for the power supply and management of the sensors as well as for the acquisition and storage of the recorded signals. Each datalogger supports 6 measurement channels and it is endowed with a 24-bit Analogical Digital converter, a GPS receiver to synchronize the acquired signals with UTC time and a battery with an autonomy of about six hours of recording. For computing the building modal masses from output-only measurements, the positions of the sensors were chosen according to the multi-polygon model (Acunzo et al. 2018), by dividing the plan of each deck into three rigid polygons (see Figure 3). The sensor layout is therefore composed of 5 measurement points for the first, the second and the third floor respectively, for a total of 30 measured acceleration components. In particular, the measurements were carried out by positioning the sensors at the level of the second, third and fourth floor. In the first two cases sensors were simply placed on the floor, whereas for the last level they were mechanically connected to the wall.

Fig. 3. Sensors layout and corresponding rigid polygons at every measured floor.

Two recordings were made with a duration of 3600s and 1800s respectively, at a sampling rate of 200 Hz. The recorded signals were processed using the "Operational Modal Analysis" module of the TestLab software, identifying the experimental modal parameters using the Polymax algorithm (Peeters and Van der Auweraer, 2005). The frequencies, damping coefficients and modal masses of the first five vibration modes are reported in Table 1, whereas the corresponding mode shapes are shown in Figure 4. From this Figure and from the modal masses (expressed as the ratio respect to the total mass of the building), it can be observed that, as it was expected on the basis of the structural conformation of the building, the first mode is mainly translational in X direction, i.e. along a direction ideally transversal to the entrance on Manara square. The second mode, on the other hand, is almost-translational in the Y direction, even if a slight rotation of the largest wing of the "U", corresponding to the nineteenth-century extension of the building, can be observed with respect to the rest of the plan. The third mode roughly corresponds to a rotation of the entire floor, even if not completely rigid. The fourth and fifth modes have a more local character, mainly involving only the wings of the U. Whereas in the fourth mode the two wings rotate in opposition to each other, in the fifth they rotate in phase. The total mass of identified modes, in x and y direction, are respectively 0.93 and 0.77 of the building mass. The correlation between the identified mode shapes are evaluated through the matrix of MAC indices (Modal Assurance Criterion) (Ewins, 2000), whose element ij represents the MAC index between mode shapes i and j.