PSI - Issue 62

Ettore De La Grennelais et al. / Procedia Structural Integrity 62 (2024) 763–772 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

768

6

6. Calibration of the FEA model based on experimental dynamic identification. Given the significance of bridge structures and the need to investigate their real behavior, especially in anticipation of potential future monitoring, as outlined in §6.3.3.5 of the Guidelines an assessment and update of the overall structural model were carried out through comparisons with static tests and dynamic identification analyses. Especially in the case of bridges with complex behavior, this activity minimizes uncertainties related to the structural model implementation. Dynamic identification, in this regard, allows for the appropriate calibration of numerical models underlying calculations, reduces model uncertainties in defining partial safety factors, and enables reliable monitoring. The results of the modal analysis aim at calibrating the model and, if necessary, detecting structural damage or malfunctions. The modal parameters obtained are representative of the dynamic behavior of the structure under real usage conditions. The test can be conducted in the absence of traffic or sometimes in the presence of moving vehicles inducing minimal operational loads. The dynamic identification serves the following purposes: Enable calibration and subsequent validation of the numerical model based on modal shapes and experimental vibration frequencies. Estimate the damping levels of the structure and assess their magnitude concerning potential damage scenarios. In this section, the primary modes of vibration of the typical span of the bridge will be studied by comparing those calculated using the FEA model used for the calculations with those experimentally observed through tests on the bridge, conducted with the aid of triaxial accelerometers. The frequencies calculated via FEA were assessed by neglecting any mass contribution related to the traffic passing over the bridge; in other words, these frequencies are attributable to the mass characteristics of the bridge's own structural and dead loads. According to the considerations in the utilized FEM model, the total mass of the modelled span of the bridge is listed in Table 2.

Table 2. Masses modelled in FEA.

mass

[ton] 123,3 200,1

G1_structural steel

G1_slab

G1_predalles

47,8

G2_non - structural steel

7,3

G2_binder G2_plants

138,2

25,5

TOTAL

542,2

6.1. Experimental campaign for the dynamic identification of the superstructure The dynamic identification test involves experimentally assessing the natural vibration frequencies of the typical span of the bridge and comparing them with those calculated using the Finite Element Model (FEM), with some uncertainties evaluated below. The test, designed by BOLINA Ingegneria, was conducted by the ANAS Laboratory in Cesano. According to the information received regarding the test performed by the Cesano Laboratory on behalf of ANAS, this test was conducted for each of the 4 spans using triaxial accelerometers, positioned as shown in Fig.5.