Finite-element (FE) model updating is a very effective procedure for determining the uncertainty parameters in a structural model and minimize differences between experimentally and analytically identified dynamic characteristics. This procedure can be practiced with manual and global/local automatic model updating procedures. Automatic model updating is very popular because of its applicability to all kinds of engineering structures to minimize differences, and its damage localization and structural health monitoring. This paper considers a historical masonry bastion to obtain its dynamic characteristics numerically with the finite-element method and experimentally with ambient vibration tests (AVTs). Sensitivity analyses for the uncertainty parameters and manual and automated model updating to minimize differences are also performed. A castle bastion located in Trabzon, Turkey is selected as an application. The FE model of structure is developed in commercial software. The first three natural frequencies are obtained between 5 and 10 Hz, and AVTs are performed under natural excitation using enhanced frequency domain decomposition (EFDD) and subspace structural identification (SSI) techniques. The first three natural frequencies are obtained between 4 and 7 Hz. The comparison of analytically and experimentally identified dynamic characteristics shows a close agreement between mode shapes, but 26% difference in natural frequencies. To minimize the difference, the FE model of the bastion is updated using manual and global/local automated updating, changing its material properties of Young’s modulus and material density. The maximum differences are reduced from 27 to 3% with manual updating, and to 0.02% with automated model updating.

Structural health monitoring (SHM) has been widely employed in civil infrastructures for a number of years. Real‐time monitoring of civil projects involves the utilization of diverse sensors. Nevertheless, accurately assessing the actual condition of a structure can pose challenges due to the existence of anomalies in the collected data. Abnormalities in this context often arise from a variety of factors, including extreme weather conditions, malfunctioning sensors, and structural impairments. The existing condition of anomaly detection is significantly impeded by this disparity. Online detection of anomalies in SHM data plays a crucial role in promptly assessing the status of structures and making informed decisions. In vibration‐based SHM, enhanced frequency domain decomposition (EFDD) is one of the most used methods in the frequency domain. The signal output obtained from EFDD also includes the frequencies of the structures, which is a holistic evaluation. The findings of frequency measurements are influenced by the presence of structural damages. Extracting damage‐sensitive characteristics from structural response has emerged as a complex task. Deep learning approaches have garnered growing interest due to their capacity to efficiently extract high‐level abstract features from raw data. Within the scope of the study, a novel approach based on anomaly detection of changes in the signal output obtained using the EFDD was developed with autoencoders in deep learning. The performance of the novel approach was examined depending on different noise ratios (0%, 0.5%, 1%, 1.5%, and 2.0%) using the Z24 Bridge dataset. In the autoencoder training model, an autoencoder model containing a 4 Conv1D layer encoder–decoder as 128 × 64 × 64 × 128 was designed. By using the signal data of the first singular values obtained with the EFDD method, grouping was made with the labels “training data (1260 pieces),” “undamaged new data (250 pieces),” and “damaged new data (320 pieces).” In addition, the upper limit of the reconstruction error was calculated as 810 using the training data in the autoencoder model. The filtered reconstruction error values obtained were compared under different noise levels. At the end of the study, it was concluded that the novel approach works effectively under different noises and can be used in anomaly detection.

This paper includes the structural condition assessment of a 164-m suspension bridge in accordance with numerical and experimental methods. A finite-element model was developed using SAP2000 software so that the assessment of the bridge could be calculated numerically. The operational modal analysis method was also used to obtain experimental dynamic characteristics. In this context: (i) numerical dynamic characteristics of the bridge were obtained and compared with the experimental dynamic characteristics, (ii) a model updating procedure was carried out to minimize the differences between the numerical and experimental natural frequencies, (iii) static and dynamic analyses of the bridge were carried out by using the initial and updated finite-element models, (iv) fatigue life assessment of the bridge was investigated, and (v) the utility of the bridge for pedestrians and vehicles was checked according to various regulations. The results showed that updating the finite-element model had a significant role to play in the static and dynamic behavior of the bridge. The results also demonstrated that the bridge was designed to be extremely safe against static, dynamic, and fatigue loads. Finally, it was concluded that the bridge investigated behaves as a girder bridge rather than a suspension bridge.