The current practice on the performance assessment of bridges, and more generally infrastructures, suffers from several deficiencies, the most-well known that of being based on visual observations of structural damage only, thus being limited on qualitative data. In depth investigations are instead needed when safety and serviceability are evaluated during structures service life, and for them, it is necessary to improve and/or establish reliable procedures and techniques to carefully assess the current health condition of structures. Experimental identification of the modal parameters through dynamic inverse methods has been one of the most interested research topics during the last years. In this thesis we used dynamic inverse methods to for structural identification. We performed dynamic tests and modal analysis techniques to calibrate detailed finite element models (FEM) of the bridges. The application of these methods to complex structures is useful for future diagnostic analyses and is innovative for the scientific community. The first complex bridge we analysed is “Rio dei Carri”. This is a prestressed concrete box girder bridge built using the cantilever system developed by the German company "Dyckerhoff und Widmann" with the use of "Dywidag” pretensioned bars. The deck is 12.50 m in width and has two spans of 57.25 m each. The bridge was designed by engineer Silvano Zorzi in the 1970s and is located in Tarvisio (Udine, Italy). To define the state of preservation of the prestressed reinforcement of the deck, we conducted an extensive in situ campaign at the end of 2022. In order to determine the main modal parameters of the bridge, we conducted dynamic test in operational conditions in October 2023. We identified ten vibration modes in the frequency range 2.8-11 Hz. The structure was studied numerically by means of EF analysis. Comparison of the three different types of prepared FEM models and the experimental methods can provide guidance to the scientific community on which modelling to choose depending on the analysis to be done.The whole research work is preparatory to the rehabilitation and retrofit that the bridge's owner, ANAS S.p.A., has planned to carry out in 2024. These results were presented at “Proceedings of IOMAC 2024 at Naple. The second bridge that analysed is called “Versa” and is a two-arch cable-stayed bridge over the Versa River, located in “Villanova dello Judrio” (Gorizia, Italy). The deck is supported by six groups of three stays which are anchored to the two steel girders of the deck and to the arches. The deck is 14.80 m in width and has one span of 110 m. The bridge was designed by Matildi + Partners and built in 2013. To determine the main modal parameters of the bridge, dynamic tests were conducted in operational conditions in October 2013 by other researchers. In the same year additional impulsive tests were carried out to identify the frequencies and the axial force on the stays. We reprocessed the results of the 2013 dynamic tests by identifying 17 vibration modes in the frequency range 1.25-7.25 Hz. The results are similar to those in 2013 published at EURODYN 2014. We calibrated a detailed finite element model of the bridge. The differences between the experimental and analytical frequencies of the M-FE model were, in all cases, small: the maximum absolute relative error was equal to 6.4%. Depending on the case a good correlation between the experimental modes and the corresponding analytical modes can be observed, with MAC values above 0.68. The 2013 studies were integrated with the estimation of the axial load on the 9+9 stays and the comparison between the theoretical design deflection from FE analysis with the static test loads. In conclusion, the application of these methods to complex bridges is useful for future diagnostic analyses and is innovative for the scientific community.

Dynamic Identification of Bridges: Two Case Studies

POZZI, ANNA MARTA
2025

Abstract

The current practice on the performance assessment of bridges, and more generally infrastructures, suffers from several deficiencies, the most-well known that of being based on visual observations of structural damage only, thus being limited on qualitative data. In depth investigations are instead needed when safety and serviceability are evaluated during structures service life, and for them, it is necessary to improve and/or establish reliable procedures and techniques to carefully assess the current health condition of structures. Experimental identification of the modal parameters through dynamic inverse methods has been one of the most interested research topics during the last years. In this thesis we used dynamic inverse methods to for structural identification. We performed dynamic tests and modal analysis techniques to calibrate detailed finite element models (FEM) of the bridges. The application of these methods to complex structures is useful for future diagnostic analyses and is innovative for the scientific community. The first complex bridge we analysed is “Rio dei Carri”. This is a prestressed concrete box girder bridge built using the cantilever system developed by the German company "Dyckerhoff und Widmann" with the use of "Dywidag” pretensioned bars. The deck is 12.50 m in width and has two spans of 57.25 m each. The bridge was designed by engineer Silvano Zorzi in the 1970s and is located in Tarvisio (Udine, Italy). To define the state of preservation of the prestressed reinforcement of the deck, we conducted an extensive in situ campaign at the end of 2022. In order to determine the main modal parameters of the bridge, we conducted dynamic test in operational conditions in October 2023. We identified ten vibration modes in the frequency range 2.8-11 Hz. The structure was studied numerically by means of EF analysis. Comparison of the three different types of prepared FEM models and the experimental methods can provide guidance to the scientific community on which modelling to choose depending on the analysis to be done.The whole research work is preparatory to the rehabilitation and retrofit that the bridge's owner, ANAS S.p.A., has planned to carry out in 2024. These results were presented at “Proceedings of IOMAC 2024 at Naple. The second bridge that analysed is called “Versa” and is a two-arch cable-stayed bridge over the Versa River, located in “Villanova dello Judrio” (Gorizia, Italy). The deck is supported by six groups of three stays which are anchored to the two steel girders of the deck and to the arches. The deck is 14.80 m in width and has one span of 110 m. The bridge was designed by Matildi + Partners and built in 2013. To determine the main modal parameters of the bridge, dynamic tests were conducted in operational conditions in October 2013 by other researchers. In the same year additional impulsive tests were carried out to identify the frequencies and the axial force on the stays. We reprocessed the results of the 2013 dynamic tests by identifying 17 vibration modes in the frequency range 1.25-7.25 Hz. The results are similar to those in 2013 published at EURODYN 2014. We calibrated a detailed finite element model of the bridge. The differences between the experimental and analytical frequencies of the M-FE model were, in all cases, small: the maximum absolute relative error was equal to 6.4%. Depending on the case a good correlation between the experimental modes and the corresponding analytical modes can be observed, with MAC values above 0.68. The 2013 studies were integrated with the estimation of the axial load on the 9+9 stays and the comparison between the theoretical design deflection from FE analysis with the static test loads. In conclusion, the application of these methods to complex bridges is useful for future diagnostic analyses and is innovative for the scientific community.
25-mar-2025
Italiano
Bridge; OMA; Dynamic Test
MORASSI, Antonino
MARCHIOLI, Cristian
Università degli Studi di Udine
File in questo prodotto:
File Dimensione Formato  
TesiAMP.pdf

accesso aperto

Dimensione 10.08 MB
Formato Adobe PDF
10.08 MB Adobe PDF Visualizza/Apri

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14242/217537
Il codice NBN di questa tesi è URN:NBN:IT:UNIUD-217537