PSI - Issue 84

Marco Civera et al. / Procedia Structural Integrity 84 (2026) 1206–1213

1209

configurations were considered, with vibration data acquired with the installed accelerometers: (i) Stationary vehicle with the engine switched off while resting on a flat and rigid surface; (ii) Stationary vehicle under the same conditions but with the engine switched on; and (iii) Moving vehicle with a known constant speed (20 km/h) on a flat and rigid surface. These tests, all conducted outside any bridge, were intended to capture the combined vibrations from the vehicle structure, the propulsion system, and the tyre-road interaction. In all three cases, a sampling frequency = 200 Hz was used. The two stationary tests lasted 5 minutes each. Notably, a more rigorous dynamic identification of the vehicle is still to be done (Gkoktsi et al., 2024). As a full-electric vehicle, it was confirmed that the engine vibrations are minimal, resulting in a considerable reduction in additive noise compared to a traditional combustion engine. Based on a first visual interpretation of the Power Spectral Densities (PSDs) of the acceleration signals for all directions and all four sensors, it seems that the EV has some own natural frequencies in the 2-5 Hz range; in future works, these will need to be isolated and removed to avoid potential misidentifications. With the engine switched on, a series of frequency peaks appears from approximately 23 Hz upwards, limiting the upper end of the bridge's natural frequency inspectable range. 3.2. The Turin (January 2025) experimental campaign The experimental test campaign in Turin and the rural surrounding area was conducted between 20-24 January 2025. During these four days, data were collected using the prototype moving platform on four different bridges, one per day. Each bridge was investigated through multiple crossings at different speeds and operating conditions, following the guidelines reported in (Massarelli et al., 2024) to investigate low and moderate speeds, directional influence, and potentially the effects of road surface roughness, as detailed hereinafter. The first bridge investigated was the Amedeo VIII bridge, a retrofitted multispan Reinforced Concrete (RC) bridge located in Turin over the Stura di Lanzo River (Fig. 2). For a detailed structural description of the infrastructure, please refer to (Civera et al., 2026). Field tests were conducted on the night between January 20 th and 21 st ; they consisted of a series of controlled passages of the iBHM+ prototype sensing platform across the bridge in both lanes and directions, as shown in Fig. 2 (b); no other vehicles were allowed to transit over the bridge during the tests. Data were acquired from all onboard sensors during the vehicle's transit over the structure. The tests were conducted under light rain (Fig. 2 (c)). Bridge passes were performed at three speed regimes, also based on (Cerda et al., 2012). In particular, a series of runs was conducted starting from the lowest possible velocity (approximately 5–10 km/h); a second set at about 20 km/h (i.e., the minimum speed maintainable with cruise control); and finally, two other sets of runs at 35 km/h and 40 km/h. In addition, two stationary measurements were performed, based on (Li et al., 2019), with the vehicle positioned at mid-length of the central span. Data were acquired for five minutes, with the engine first switched off and then switched on, in both cases with passengers on board. The second case study investigated the Regina Margherita bridge, a prestressed RC arch bridge over the Po River in Turin (Fig. 3 (a)). Notably, this bridge is also equipped with a permanent SHM system for dynamic monitoring; a description is provided in (Civera et al., 2025). As for the Amedeo VIII bridge, a set of controlled runs was performed on this second bridge, investigating five speed regimes (5, 10, 20, 40, and 45 km/h). However, in this particular case, traffic was interrupted on a single carriageway (the downstream one, highlighted in Fig. 3(b)), and all measurements were consequently conducted by driving the iBHM+ vehicle back and forth along the temporarily closed three lanes, with normal traffic on the upstream side. Stationary measurements were also performed, with the vehicle positioned at mid-length and at ¼ of the main (central) span, with the engine both switched off and on, for a total of 4 measurements, each lasting 5 minutes. In addition to these tests, traditional (i.e., contact) accelerometers were deployed on the investigated side of the structure at the positions shown in Fig. 3 (c). These data, recorded synchronously, will serve as a direct basis for benchmarking the acquisitions and dynamic identification results obtained with the mobile platform. The tests lasted the whole day of January 21 st .