PSI - Issue 84

Domenico Liberatore et al. / Procedia Structural Integrity 84 (2026) 1151–1158

1154

overlap among the acquired data. Data acquisition was carried out at multiple distances, with scanning positions arranged to be as orthogonal as possible to the bridge surface. The lasers were inclined at a maximum incidence angle of 40°, to guarantee adequate spatial resolution, including the peripheral areas of the survey. The resulting cloud of points (Fig. 2a) is subsequently processed to extract the 2D representation in AutoCAD of the two prospects in elevation (North and South) and three cross-sections at the height of 7, 12 and 23 m. In addition, a photogrammetric rectification process is performed by comparing in situ measurements with the metric data obtained. A camera calibration and optical distortion correction are conducted, using the Acca FOTUS software, enabling the generation of an accurate photoplan and the derivation of more detailed geometric and dimensional information of the bridge. 3.2. Ambient vibration analysis The ambient vibrations measurements on Pons Fabricius were carried out on July 11 th , 2025, using Tromino®, a single-station digital tromograph operable in both passive and active modes, developed by Micromed S.p.A. The tool is capable of recording ambient vibrations over a wide frequency range with high-resolution velocimetric and accelerometric channels.

Fig. 3. Velocity amplitude spectra of the 10 measurements.

As shown in Fig. 2b, the instrument was placed on the roadway surface, with its axis aligned along the longitudinal direction of the bridge, from Tiber Island towards the bank of the Tiber, corresponding to the X-axis of the numerical model. A sampling frequency of 128 Hz was adopted, and each measurement had a duration of 16 minutes. The locations of the measurements were chosen based on the expected deformation shapes of the bridge. Assuming a sinusoidal first vibration mode, the first point was located at the mid-span; for the second vibration mode, points at one-quarter and three-quarters of the bridge length were chosen. In addition, points at the bridge ends were considered to account for the interactions between the bridge, the retaining walls on the bank of the Tiber and Tiber Island. Due to the continuous pedestrian traffic during the acquisition period, measurements were performed on the lateral sidewalks of the bridge. A total of ten measurements were carried out, along with one free-field measurement performed at the central area of Tiber Island, far from vehicular and pedestrian traffic. The ambient vibration records are processed using the Fast Fourier Transform (FFT) to obtain amplitude spectra, capture directional effects and identify the natural frequencies. Frequencies below 1 Hz and over 25 Hz are excluded due to sensor limitations and lack of significant peaks.