In the application of structural health monitoring, the measured data might be temporarily or permanently lost due to sensor fault or transmission failure. The measured data with a high data loss ratio undermine its ability for modal identifications and structural condition evaluations. To reconstruct the lost data in the field of structural health monitoring, this study proposes a deep convolutional generative adversarial network which includes a generator with encoder–decoder structure and an adversarial discriminator. The proposed generative adversarial network model needs to understand the content of the complete signals, as well as produce realistic hypotheses for the lost signals. Given the data stably measured before the occurrence of data loss, the generator is trained to extract the features maintained in the data set and reconstruct lost signals using the responses of the remaining functional sensors alone. The discriminator feeds back the distinguished results to the generator to improve its reconstruction accuracy. When training the model, the reconstruction loss and the adversarial loss are employed to better handle the low-frequency features and high-frequency features of the signals. The effectiveness and efficiency of the proposed method are validated by two case studies. As the number of training epoch increases, the reconstructed signals learn the features from low-frequency to high-frequency, and the amplitude of the reconstructed signals gradually increases. It can be seen that the final reconstruction signals match well with the real signals in the time domain and frequency domain. To further demonstrate the applicability of the reconstructed signals in data analysis, the reconstructed acceleration data are used to accurately identify the modal parameters in the numerical case, and the vehicle-induced responses are precisely decomposed from the reconstructed strain data in the field case. Finally, the reconstruction capacity is also investigated with the different numbers of the faulted strain gauges.

One of the main challenges for structural damage detection using monitoring data is to acquire features that are sensitive to damages but insensitive to noise (e.g. sensor measurement noise) as well as environmental and operational effects (e.g. temperature effect). Inspired by the capabilities of deep learning methods in representation learning, various deep neural networks have been developed to obtain effective damage features from raw vibration data. However, most of the available deep neural networks are supervised, resulting in practical difficulties owing to the lack of damage labels. This article proposes a damage detection strategy based on an unsupervised deep neural network, referred to as deep convolutional denoising autoencoder, which accepts multi-dimensional cross-correlation functions as input. The strategy aims to extract damage features from field measurements of undamaged structures under the influence of noise and temperature uncertainties. In the proposed strategy, cross-correlation functions of vibration data are first calculated as basic features; then deep convolutional denoising autoencoder is developed to reconstruct cross-correlation functions from their noise-corrupted versions to extract desired features; exponentially weighted moving average control charts are finally established for these features to identify minor structural damages. The strategy is evaluated through a numerical simply supported beam model and an experimental continuous beam model. The mechanism of deep convolutional denoising autoencoder to extract damage features is interpreted by visualizing feature maps of convolutional layers in the encoder. It is found that these layers perform rough estimations of modal properties and preserve the damage information as the general trend of these properties in multiple extra frequency bands. The results show that the proposed strategy is competent for structural damage detection under the exposed environment and worth further exploring its capabilities in applications of real bridges.

Structural deformation is an important consideration in the health monitoring of bridges, and its dependence on temperature variations is quite complex. Based on field measurements performed for an operational cable-stayed bridge, the proposed study investigates mechanisms of thermally induced variations in girder length and mid-span deflection through plane geometric and finite element analyses. The objective of this study is to understand the behaviour of such bridges over annual and diurnal cycles. It has been observed that the girder length and mid-span deflection of a cable-stayed bridge exhibit different modes of the temperature–response correlation. Thermally induced changes in girder length are solely governed by the average girder temperature, and its annual variation in amplitude is significantly larger compared to the diurnal variation. However, thermally induced mid-span deflections are simultaneously influenced by the cable temperature and average girder temperature, and these do not vary monotonously with temperature, thereby resulting in nearly equal variation amplitudes over both annual and diurnal cycles. Temperature-induced deformations of a cable-stayed bridge could well be approximated through multiple linear superposition of thermal-expansion effects of individual components. Besides thermal-expansion coefficients of structural materials, the temperature dependency of mid-span deflection of a symmetrical twin-tower cable-stayed bridge is closely related to the ratio of tower height above the deck to central span of the girder as well as span ratio of the side span to central span. The proposed simplified formulae to estimate the sensitivities of temperature effects could be readily extended to other cable-stayed bridges with different geometric arrangements, thereby providing valuable insights into thermally induced deformation of such bridges.

A new framework for bridge-deflection estimation through inclinometer data is presented in this paper. Whether the structure is damaged or not, deflection could be precisely estimated via the finite-element model (FEM) combined with the partial least-square regression (PLSR) method. In addition, damage localization is achieved at the same time. First, the deflection of a damaged structure is equivalent to that of an undamaged structure but subjected to special virtual loads. Second, the basis functions used in regression are calculated by the FEM of the undamaged structure rather than pure mathematical derivations. Third, the PLSR model is built on the basis of these basis functions and measured inclination data. By solving the PLSR equation, all the nodal inclinations and corresponding nodal loads of the structure could be obtained despite a limited number of measured nodes. Finally, the rotation-displacement transfer matrix is derived to reconstruct deflection from estimated inclinations, and the changes in nodal loads are served as damage indicators. To verify its effectiveness, numerical simulations considering various damage scenarios and different noise levels are performed. The results indicate that the proposed method is quite accurate and reliable in deflection estimation, and it could provide rough damage localization as well, which reveals a great potential in the field of structural health monitoring.