Seismic Performance Analysis of Bridges with Isolation Devices Enhanced by Hybrid Dynamic Substructuring

The Seismic Performance Analysis of Bridges (SPAB) constitutes one of the biggest challenges for structural and civil engineers. In fact, the handling of these design problems requires a deep knowledge of structural behavior and a huge expertise with numerical and analytical tools necessary to perform advanced Finite Element (FE) simulations including dynamic and probabilistic aspects. Within the scope of SPAB, this thesis proposes the analysis of complex bridges assisted by the profitable well-known method of Dynamic ubstructuring (DS), advanced model updating strategies, fully probabilistic approaches and innovative time integration algorithms. SPAB includes the evaluation of several nonlinear behaviors inside the structural components and the quantification of benefits generated by safety systems such as isolation devices. As a result, in order to highlight the main advantages of a well designed isolation system, most of the cases analyzed include the comparison between non isolated and isolated configurations. In greater detail, four different bridges have been analyzed and will be presented in this thesis. First, the Rio Torto highway viaduct, an existing Reinforced Concrete (RC) viaduct on the A1 Italian highway between Florence and Bologna. The structure has been investigated at the laboratory of the Joint Research Center in Ispra (VA) by means of Hybrid Simulations (HSs). The set of 1 : 2.5 scaled substructures included two RC frame piers and the isolation system. The critical issues of the structure due to the complexity of the geometry and the awfulness was the presence of poor seismic details characterized by plain steel rebars. Owing to lack in knowledge for this type of rebars, tests were needed to analyze the seismic response in the as built configuration and to evaluate the effectiveness of a seismic retrofitting designed with a traditional Concave Sliding Bearings (CSBs) system. Then, a typical RC bridge with an innovative prototype of Concave Sliding Bearing (CSB) has been tested at the EUCENTRE Tress Laboratory in Pavia (PV) through HSs. The set of Physical Substructures (PSs) included a 1 : 2 scaled RC box section pier and a full-scale CSB. The prototype was characterized by an asymptotic relation between friction coefficient and load rate. All the benefits of the DS were exhibited during the test; in fact, to exploit the actual potentiality of the isolation system, even with the low speed of the test, the restoring force coming from the CSB was numerically corrected at each time step. Furthermore, a short-medium span Steel Concrete Composite Bridge made with Hot rolled I-girders (SCCBH) has been investigated. The SCCBH is an example of structural optimization; in fact, it combines both economic and functional benefits deriving from the reduction of in site works, e.g. welding, and short construction time. In particular, The novelties were threefold: i) the testing of a novel connection between a steel I-girder and a Concrete Cross Beam (CCB); ii) the development of a novel mechanical model for this connections; iii) the application of the Performance Based Earthquake Engineering (PBEE) to SCCBH. The experimental campaign has been performed on six 1 : 2 scaled substructures, representing a deck subassembly, tested in both longitudinal and transverse loading directions. Finally, a simulation-based reliability assessment of a complex cable-stayed foot/cyclic bridge located close to the sea and equipped with dynamic viscous dampers was performed. The scope was to investigate the benefits of Circular Hollow Section (CHS) structural members for this type of structure when erected in an aggressive environment. A FE model of the structure has been validated, and then used to perform a probabilistic time dependent analysis. Therefore, two corrosion models, i.e. general and localized, capable of evaluating the reduced load bearing section were implemented; and appropriate probability distribution functions were assigned to input model parameters to evaluate the response of the facility during its service life. As a result, the time dependent probabilities of failure have been evaluated and compared with the codes prescriptions.