PSI - Issue 84

Tommaso Pivetta et al. / Procedia Structural Integrity 84 (2026) 1286–1293 T. Pivetta et al. / Structural Integrity Procedia 00 (2026) 000–000

1289

4

The accelerometers were set to acquire data at a sampling frequency of 64 Hz, and the recorded signals were processed by removing the trend and applying a band-pass filter. The deflection response of the deck at the midspan was instead measured using radar systems.

a)

b)

Fig. 4. Examples of signal acquisition during the tests: (a) Accelerations (b) Deflection at midspan.

3. High-fidelity model: development and calibration This section presents the calibration process of the numerical model of the bridge. First, the finite element model developed from the available design documentation and survey data is described. Then, the calibration strategy adopted to align the numerical response with the experimentally measured dynamic and static parameters is illustrated. 3.1. FEM model Starting from the available original design drawings and the recent geometric survey, a high-fidelity finite element (FE) model of the first span was developed using Midas Civil NX (MIDAS Information Technology Co., 2025). In order to obtain an adequate agreement between the measured and numerical dynamic parameters, the main girders and the cross-beams were modelled using beam elements, while plate elements were adopted for the deck slab. Material properties (mainly density and Young’s modulus) were initialized according to the outcomes of the materials characterization campaign, whereas the boundary conditions were defined based on the original design documents. Fig. 5 provides a representation of the FEM model of the deck. The modelling activity focused exclusively on the deck of the first span of the bridge, as this portion of the structure was the object of the monitoring and testing campaign.

Fig. 5. Finite element model of the bridge deck.

Since the objective of the study was the calibration of the dynamic response under operational conditions, only linear elastic materials behavior was assumed. Table 2 summarizes the initial values of the Young’s Modulus (E c ), as derived from the available documentation, whereas the unit weight of all concrete components was set equal to the conventional value of 25.00 kN/m 3 .