PSI - Issue 84

Donato Fiore et al. / Procedia Structural Integrity 84 (2026) 417–424

423

= √ 2

(1)

Therefore, a reduction in frequency of the mode shapes would result not so much from a reduction in stiffness, but rather from an increase in mass. Comparing the corresponding modes (Tab. 1), it is observed that the frequency of the first three horizontal modes of the arch - two symmetric and one semi-symmetric, with frequencies of 0.79, 1.53, and 2.48 Hz respectively - in the complete bridge assume frequencies equal to 0.55, 0.73, and 0.94 Hz. The first two vertical frequencies change from 0.85 and 1.48 Hz in the arch to 1.27 and 1.18 Hz in the complete bridge. This comparison demonstrates that for transverse oscillations, which are of greater relevance for seismic verification, the increase in mass prevails in the dynamic response of the complete bridge. Conversely, for oscillations in the plane of the bridge, the arch/deck interaction determines not only a significant increase in mass but also a structural interaction that increases the stiffness of the structural system for certain modal forms. It is also noted that the local modes, which affect the individual support piers of the deck, are identified at significantly higher frequencies, which increases the complexity of monitoring of individual structural elements using dynamic analysis methods. 5. Conclusions In conclusion, the following points were defined for the Bisantis bridge: • It has a main structure that appears symmetrical, but exhibits a partially asymmetrical response due to the asymmetry of the road connections at its ends; • It has clearly identifiable natural frequencies, in the range of 1–5 Hz for the most significant global modes, with some frequencies associated with local modes exceeding 10 Hz; however, these local modes are difficult to identify either theoretically (FEM model) or experimentally; • The monitoring system (instruments and acquisition system) and the structure show responses that are not perfectly constant over time, as typically expected, but vary within ±6–7% of the mean value. This observation is fundamentally important both for defining threshold values (alert and alarm levels) and for evaluating a potential anomaly condition: an anomaly occurs when the deviation from known frequencies exceeds 10%; • The identification of the natural frequencies of the Bisantis Bridge is possible under current road traffic conditions within the limits discussed above. The use of road traffic means that the first natural modes may not produce the highest acceleration values of the bridge, depending on the type (speed, weight, geometry) of the passing load. The interferometric system showed good results in accordance with accelerometric data: the system can be used as a rapid tool for a preliminary yet accurate evaluation in SHM.