PSI - Issue 84

Michele Morici et al. / Procedia Structural Integrity 84 (2026) 89–96

92

been installed on the above-mentioned beams, three of them on beam #2 and the remaining two on beam #5. In detail beam #2 is equipped with three tiltmeters located at 1/3, 2/3 and nearly the end of its length, while beam #5 has tiltmeters only at 1/3 and 2/3 of its span. Finally, three strain gauge rods have been installed at 1/3, midspan and 2/3 of the length of beam #2. The SHM system of the bridge is completed by additional contact sensors installed at the base of the pier (four strain gauge rods) and on its cap (a thermocouple and a biaxial tiltmeter). Such sensors are not shown in Figure 1 due to space constraints. Bridge monitoring began at the end of May 2024. From June until July 18th, observations were conducted under traffic-free conditions, as the bridge remained closed for maintenance work.

span#1

span#2

TC1-s

TC5-s

Lh

# beam number

TC2-b1

TC3-b1

#1

Lv-b1

SG1-b2 SG2-b2

SG3-b2

#2

horizontal LVDT strain gauge rod vertical LVDT biaxial tiltmeter

T4 b2x

T2 b2x

T3 b2x

#3

TC1-b2

Lv-b2

TC5-bt

#4

pier

#5

abutment A

abutment B

T1-b5x

T3-b5x

#6

Lv-b5

thermocoupl e above the thermocouple under the slab

TC2-b6

TC3-b6

Lh

TC3-s

TC4-s

TC6-s

y

Global reference

z

x

system

Fig. 1. Schematic view of the sensors installed in the SHM system highlighting those installed on the deck of the bridge.

3. Preliminary results This section presents selected preliminary results from the monitoring system. Figure 2 illustrates the time series of rotations and vertical displacements measured on beam #5 at different locations between January 2025 and May 2025. With reference to Figure 1, the data corresponds to sensors T1-b5y, T3-b5y, and LVDT-b25. In Figure 2, blue curves represent the raw measurements, while red curves denote the filtered signals obtained using a one-dimensional median filtering procedure. As shown in Figure 2, the recorded signals are characterized by a certain noise level, due to traffic load. The daily variations are visible to be mainly attributed to the temperature fluctuations occurring throughout the day. Under operational conditions, displacement responses (Figure 2c) become more pronounced than rotational responses, with decreasing displacement values indicating downward deflections. Furthermore, until April, when ambient temperatures remain relatively low (approximately below 20 °C), rotations T1‑b5y (near the central pier) exhibit higher fluctuations compared to those observed from May onward. These observations may indicate a greater sensitivity of the bridge response to deformation at lower temperatures. In contrast, rotations T3‑b5y exibit only very slight oscillations, indicating that rotations near abutment B are significantly restrained. A possible explanation could be a partial reduction in the functionality of the bearings and the expansion joint at abutment B, potentially associated with the presence of debris. As illustrated in Figure 3, debris accumulation at the supports (Figure 3a), together with the closure of the expansion joint (Figure 3b), may limit the bridge expansion towards abutment B, especially under warm conditions. An additional contributing factor may be associated with the deck-to-pier connection near the central pier. High temperatures, combined with debris accumulation at abutment B, may locally modify the boundary conditions, leading to a structural response more consistent with a continuous deck rather than a simply supported configuration. Figure 4 depicts the relationship between rotations θ (in degrees) or vertical displacements D and temperature (in °C) at different monitored locations. The regions approximately located above 20°, correspond to thermocouples installed above the deck slab and identify temperature ranges potentially affected by direct solar exposure. These areas highlight the presence of differential heating effects with respect to thermocouples installed beneath the slab. In Figs. 4a) –d)–g)–j)–m), representing tiltmeter T1- b5y, and Figs. 4c) –f)–i)–l)–o), showing LVDT LV-b5, a general trend can be observed in which both rotation and deformation measures, and , increase with rising temperature. The rate of increase becomes less pronounced at higher temperatures, suggesting a nonlinear correlation.