**1. Introduction**

Long-span bridges, such as cable-stayed and suspension bridges having considerably different structural system compared with other structures are one of the main components in the transportation network in a city/state. Therefore, the governments make sizeable investment

Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons

both for their construction and maintenance costs as wells as rehabilitation and retrofit projects. Considering such crucial functions of them and various investments for improving their structural performance, it is clear that comprehensive investigations on long-span bridges are necessary before and after its construction since they fall outside the scope of general bridge codes.

The most comprehensive study for St-Id was carried out by the ASCE SEI Committee on Structural Identification of Constructed Systems [1]. In addition to detailed knowledge for this process given in this study, a set of case studies are presented to illustrate the application of St-Id to building and bridge structures. Six chapters are introduced to properly comprehend St-Id concept. The scope of these sections is related to modeling, experimentation, data processing, comparison of model and experiment and decision-making. As schematically shown in **Figure 1**, St-Id consists of six steps: (i) Visualization and conceptualization (ii) Priori modeling (iii) Experimentation (iv) Data analysis (v) Calibration (vi)

Structural Identification (St-Id) Concept for Performance Prediction of Long-Span Bridges

http://dx.doi.org/10.5772/intechopen.71558

45

In addition to theoretical aspects, St-Id is widely utilized from estimating structural condition evaluation, rehabilitation to SHM system improvement and optimization of scaled and existing/new constructed long-span bridges under different loading and environmental conditions. A St-Id study on a newly constructed arch bridge in China was conducted in order to make FE model updating of the bridge through certain data-driven methods [6]. They demonstrated that the updated numerical model could be reliably adopted for the detailed structural analysis of arch bridges. Similar study was made in [7]. They proposed a method of frequency response function parameter identification for full-scale bridge using field-testing data. The developed approach to the model updating/calibration of the bridge is indicated to be confidently considered for FE model calibration of full-scale bridges. For various objectives of structural identification, many studies were carried out from the numerical model calibration to SHM system modification/improvement in [8–13]. In these studies, either all or certain steps of St-Id concept are aimed to be identified especially for long-span cable-supported

The Bosphorus Bridge as shown in **Figure 2** also called the 1st Bosphorus Bridge is one of the three long-span bridges in Turkey, which provides a connection between two continents of

Utilization.

bridges.

**Figure 1.** Layout for structural identification (St-Id).

Over the last decades, the engineers and researchers have focused on both experimental and analytical studies taking the complexity and non-linear characteristics of long-span cablesupported bridges into consideration. For analytical studies, these structures are modeled numerically by finite element model (FEM). With utilizing FEM method, modal analysis, buckling analysis and earthquake-induced analysis of complex structures are readily carried out. Although this sophisticated analytical modeling is a powerful tool for simulating three dimensional local and global behavior of long-span bridges, it does not reflect the real structural behavior owing to uncertainty in idealization and assumption, such as in material model and analysis method. Therefore, only FE model is not enough for fully understanding the response and behavior of this type of large-scale structures.

Experimental field studies of long-span cable-supported bridges give reliable consequences for the current performance of them. Therefore, this effort is relatively significant to compare the outcomes from FEM analysis with those obtained from experimental field studies; however, this analysis could not be possible many times, especially for long-span cable-supported bridges. Additionally, data analysis and management are necessary for interpretation of collected experimental data. For this objective, structural health monitoring (SHM) is proposed as a complementary experimental technique, including structural condition evaluation, model development and calibrating, and real-time monitoring. With the major leaps in sensing and information technology, utilizing SHM to properly assess new designed and existing structures is growing interest worldwide though a standard approach for SHM has not been available yet.

In an effort to deal with the incompatibility between measured and simulated response of long-span cable–supported bridges and make an overall condition assessment of these bridges, structural identification (St-Id) concept presents an integrated approach. Structural identification is the process of developing calibrated/updated analytical model (mathematical or geometric) of structures from experiments and observations for reliable performance evaluation, simulation and decision-making. Structural identification (St-Id) and system identification (Sys-Id) are two concepts that are closely related to each other. Therefore, St-Id is a result of translation and application of system identification to mechanical and civil structural system [1]. This concept was first applied to engineering mechanics [2] and to civil-structural engineering [3]. Although many efforts are made to implement St-Id to structural engineering field, applications of this concept have not been wide-spread in civil engineering practice yet [4]. Depending upon developments in sensing and information technology, St-Id concept becomes more popular for confidential performance assessment and simulation of structures [5]. For different aspect of the St-Id concept, a number of studies were found in literature. The key point in these studies is that structural health monitoring (SHM) is utilized for experimental step of St-Id.

The most comprehensive study for St-Id was carried out by the ASCE SEI Committee on Structural Identification of Constructed Systems [1]. In addition to detailed knowledge for this process given in this study, a set of case studies are presented to illustrate the application of St-Id to building and bridge structures. Six chapters are introduced to properly comprehend St-Id concept. The scope of these sections is related to modeling, experimentation, data processing, comparison of model and experiment and decision-making. As schematically shown in **Figure 1**, St-Id consists of six steps: (i) Visualization and conceptualization (ii) Priori modeling (iii) Experimentation (iv) Data analysis (v) Calibration (vi) Utilization.

In addition to theoretical aspects, St-Id is widely utilized from estimating structural condition evaluation, rehabilitation to SHM system improvement and optimization of scaled and existing/new constructed long-span bridges under different loading and environmental conditions. A St-Id study on a newly constructed arch bridge in China was conducted in order to make FE model updating of the bridge through certain data-driven methods [6]. They demonstrated that the updated numerical model could be reliably adopted for the detailed structural analysis of arch bridges. Similar study was made in [7]. They proposed a method of frequency response function parameter identification for full-scale bridge using field-testing data. The developed approach to the model updating/calibration of the bridge is indicated to be confidently considered for FE model calibration of full-scale bridges. For various objectives of structural identification, many studies were carried out from the numerical model calibration to SHM system modification/improvement in [8–13]. In these studies, either all or certain steps of St-Id concept are aimed to be identified especially for long-span cable-supported bridges.

The Bosphorus Bridge as shown in **Figure 2** also called the 1st Bosphorus Bridge is one of the three long-span bridges in Turkey, which provides a connection between two continents of

**Figure 1.** Layout for structural identification (St-Id).

both for their construction and maintenance costs as wells as rehabilitation and retrofit projects. Considering such crucial functions of them and various investments for improving their structural performance, it is clear that comprehensive investigations on long-span bridges are necessary before and after its construction since they fall outside the scope of general bridge

Over the last decades, the engineers and researchers have focused on both experimental and analytical studies taking the complexity and non-linear characteristics of long-span cablesupported bridges into consideration. For analytical studies, these structures are modeled numerically by finite element model (FEM). With utilizing FEM method, modal analysis, buckling analysis and earthquake-induced analysis of complex structures are readily carried out. Although this sophisticated analytical modeling is a powerful tool for simulating three dimensional local and global behavior of long-span bridges, it does not reflect the real structural behavior owing to uncertainty in idealization and assumption, such as in material model and analysis method. Therefore, only FE model is not enough for fully understanding the

Experimental field studies of long-span cable-supported bridges give reliable consequences for the current performance of them. Therefore, this effort is relatively significant to compare the outcomes from FEM analysis with those obtained from experimental field studies; however, this analysis could not be possible many times, especially for long-span cable-supported bridges. Additionally, data analysis and management are necessary for interpretation of collected experimental data. For this objective, structural health monitoring (SHM) is proposed as a complementary experimental technique, including structural condition evaluation, model development and calibrating, and real-time monitoring. With the major leaps in sensing and information technology, utilizing SHM to properly assess new designed and existing structures is growing interest worldwide though a standard approach for SHM has not been

In an effort to deal with the incompatibility between measured and simulated response of long-span cable–supported bridges and make an overall condition assessment of these bridges, structural identification (St-Id) concept presents an integrated approach. Structural identification is the process of developing calibrated/updated analytical model (mathematical or geometric) of structures from experiments and observations for reliable performance evaluation, simulation and decision-making. Structural identification (St-Id) and system identification (Sys-Id) are two concepts that are closely related to each other. Therefore, St-Id is a result of translation and application of system identification to mechanical and civil structural system [1]. This concept was first applied to engineering mechanics [2] and to civil-structural engineering [3]. Although many efforts are made to implement St-Id to structural engineering field, applications of this concept have not been wide-spread in civil engineering practice yet [4]. Depending upon developments in sensing and information technology, St-Id concept becomes more popular for confidential performance assessment and simulation of structures [5]. For different aspect of the St-Id concept, a number of studies were found in literature. The key point in these studies is that structural health monitoring (SHM) is utilized for experimental

response and behavior of this type of large-scale structures.

codes.

44 Bridge Engineering

available yet.

step of St-Id.

equivalent sectional properties of structural components. On the other hand, multi-scale modeling becomes very significant upon making analysis for local structural component. This modeling technique also provides that different element types, beam, solid, shell, and truss, are used together to establish 3D full-scale bridge FE model. Accordingly, both modeling techniques are utilized for different goals, and generally, they have been used

Structural Identification (St-Id) Concept for Performance Prediction of Long-Span Bridges

http://dx.doi.org/10.5772/intechopen.71558

47

The modal analysis of structures is a powerful tool for earthquake excitation analysis of structures. Through this analysis, the response of structures to dynamic input can be estimated and certain outcomes related to dynamic inputs can be explained [17]. For large-scale bridge structures with different large size of structural component, such as main deck, tower etc., the mode shapes may show which component dominates the dynamic response of long-span

The bridge's structural components of the tower, the main deck, the portal beams, and approach span are modeled as equivalent frame element corresponding their mechanical

together for structural analysis of bridge structures.

**Figure 3.** Sectional properties of the Bosphorus Bridge [17].

bridges [14–16].

**Figure 2.** General views of the Bosphorus Bridge.

the Asian and the European. When opened to traffic in 1973, the bridge was classified as the 4th longest suspension bridge in the world according to its main span. The bridge serves as vital link on the Motorway-1 (O1) connecting the city center of Istanbul. Significant part of heavy traffic of Istanbul has been carried from the bridge along with the Fatih Sultan Mehmet Bridge named the 2nd Bosphorus Bridge located on the northern side of the Bosphorus Bridge.

Taking the efficiency of St-Id into account, this approach is utilized for the Bosphorus Suspension Bridge and the results are presented in this study. Each step of the concept is handled to be conducted for the bridge. Along with establishing the numerical model of the bridge to obtain its analytical modal frequencies, the experimental study is conducted using SHM data collected during the extreme wind event. Based on the discrepancy between FEM and experimental results, model verification/calibration is performed. The calibrated model is then used for utilization of multi-point earthquake analysis of the bridge. The outcomes from this study illustrated that St-Id concept is a robust methods to properly predict the structural performance of the Bosphorus Bridge.
