Structural Control & Health Monitoring最新文献

In the process of infrastructure construction in recent decades, there exist millions of bridges in service that need safety inspection for performance assessment. Currently, computer vision and deep learning-based surface damage detection methods can achieve classification and localization of damages at the image level, but achieving precise localization in three-dimensional space is more challenging. To overcome aforementioned limitations, this study proposes a three-stage method of bridge surface damage detection and localization based on three-dimensional (3D) reconstruction. In stage 1, the UAV flight path planning of the bridge is carried out, and the 3D reconstruction model of the bridge is formed based on the structure from motion (SfM) algorithm. In stage 2, you-only-look-once version 7 (YOLOv7) network is adopted to detect multiple damages, and scale invariant feature transform (SIFT) detector is used to match the identical damage in image level. In stage 3, based on solution of epipolar geometric constraint, the matched damage was mapped to the 3D model, and the 3D damage localization is realized and visualized. The quality of the 3D model has been analyzed, and it is recommended that inspection distance is determined at 20 m. Moreover, the reconstruction model of bridges achieves centimeter-level positioning accuracy, and the positioning accuracy of damage reaches the meter level. The mapped model effectively showcases surface damages, providing bridge owners with intuitive insights.

The parameters of most conventional passive dampers are constant, which may not sufficiently satisfy the different energy dissipation capacity demands of the structure under different load conditions. The development of passive dampers with variable performances has become an emerging and vital trend in energy dissipation technologies and smart structures. This study proposes a novel passive viscous fluid damper with variable performance under different deformation levels called an asynchronized parallel double-stage viscous fluid damper (APDVFD). It is expected to exhibit an asynchronized double-stage working mechanism based on a variable annular gap. In the first stage, only the primary piston provides the damping force. When the deformation reaches a preset value, the primary and second pistons work in parallel, providing a damping force concurrently. Circular orifices are adopted for the piston head to provide a sufficient damping force. The double-stage operating mechanism and fatigue performance of the APDVFD were validated and investigated through a full-scale experiment with 46 load cases. Based on these, a theoretical model capable of predicting the hysteretic behavior of the APDVFD was developed and validated against test data.

The temperature dependence of the internal structure and elastic properties of concrete is revealed by subjecting concrete to different heating conditions. The variation trend of wave velocity in the concrete medium with temperature is analyzed through ultrasonic properties. The decrease in cement matrix modulus and the increase in crack density in concrete are the main factors leading to a decrease in wave velocity. The changes in the composition of the concrete matrix after dehydration are obtained using a thermal decomposition model. Based on the effective medium model, the calculation results of the effective modulus at different temperatures are presented, with a focus on analyzing the influences of the temperature-dependent changes in the matrix elastic properties and the randomly distributed cracks on the effective modulus. The experimental tests and the presentation of the model results indicate a relatively satisfactory agreement, thereby verifying the reliability of the models. The results of this study can explain the basic propagation mechanism of waves in concrete and have promising applications in the ultrasonic testing of thermal damage to cement-based materials.

The negative stiffness damper (NSD) has emerged as a promising passive vibration control device for cable structures due to its simplicity and effectiveness. However, uncertainties stemming from both internal and external environmental factors can potentially compromise the NSD’s performance in real-world applications, posing risks to cable safety. In response, this paper conducts a robustness evaluation on an integrated cable-NSD system, taking into account various potential uncertainties. Specifically, the uncertain parameters are described by interval variables. Consequently, an interval model is constructed to delineate the boundaries of cable dynamic responses when subjected to these uncertainties. The model’s accuracy is validated against experimental results. Subsequent simulations involve assessing interval responses for both single- and multimode cable vibrations under varying uncertainties. Finally, the NSD’s robustness concerning cable vibration control is evaluated using the model, which incorporates the first-passage theory. This analysis delves into the relationships among confidence levels, performance measures, and the variation range of uncertainties. The results indicate that for single-mode vibration control, there is a 90% confidence level that the damping ratio reduction remains within 10%. As for multimode vibration control, a 90% confidence level is established that the amplification falls within 17%.

The nonlinear energy sink (NES) has drawn increasing research attention in recent years. In this study, we investigated a novel disc spring-based vertical NES device capable of counterbalancing payload gravity and providing a combination of linear and cubic restoring forces. The vertical NES device is used to mitigate vibration responses in vertically flexible structures caused by human or machinery-induced actions. To simplify the NES controlled system, a two-degree-of-freedom model was used to account for harmonic forces acting on the primary structure. Using the complexification-averaging (CX-A) method, we obtained the frequency-amplitude solution and subsequently derived the force transmissibility (T_f). Parametric comparisons between the NES and traditional tuned mass damper (TMD)-controlled systems were performed by comparing the T_f curves and numerically verifying the theoretical solutions. Results confirmed the accuracy of the theoretical force transmissibility solutions and affirmed the advantage of the NES system by overcoming the limited operating frequency bandwidth of the traditional TMD system. The inclusion of a linear stiffness term in an NES system has been shown to enhance the vibration control performance within the resonant frequency bandwidth. In addition, the time history energy analysis for the NES and TMD systems was conducted. The energy flow analysis elucidated the broad operating bandwidth of the NES system. This finding suggests that the vibration energy flow to the attachment in the NES system is irreversible, thereby preventing amplification of the response in the primary structure, even when the system is mistuned.

Most of vision-based methods for structural damage detection rely on supervised learning, requiring a substantial number of labeled images for model training, which is labor-intensive and time-consuming. To address these challenges, this study introduces a vision-based structural anomaly detection and localization approach using unsupervised learning and reverse knowledge distillation. The proposed model incorporates a teacher model, a student model, and a trainable one-class bottleneck embedding module. The asymmetrical architecture of the teacher and student models forms an encoder-decoder structure for parameter transfer and feature extraction. The student network receives a specific embedding from the teacher network as input and target, facilitating the recovery of multiscale information from the teacher. Training images only contain the undamaged structures, and the teacher model, a pretrained model, instructs the student model to remember their undamaged features to detect and localize damages in unseen testing images. Through experiments, including a comparison among five candidate backbones for pretrained teacher models based on the residual network and testing across various structural damage types, the optimal model is identified, demonstrating good performance in both anomaly detection and localization. Furthermore, the model’s generalization performance is thoroughly validated, confirming its efficacy across diverse scenarios.

In addressing the problems of delayed detection, inefficient identification, and coverage blind spots in optical fiber-based pipeline leakage detection within pipe galleries, this study proposes a leakage detection strategy utilizing distributed optical fiber temperature measurement technology. Finite element method is employed to analyze the temperature influence radius around the pipeline leakage hole and a model test is conducted as a validation. The results show that the temperature field image at the leakage site is elliptical, influenced by temperature differences between ambient and liquid. An increase in this temperature difference accelerates changes in the temperature field’s range. Adjustments to the optical fiber’s winding angle and pitch demonstrated that an optimal pitch is 1/24 of the pipeline’s length, with a 45° winding angle. This configuration maximizes the optical fiber’s distribution in detection while maintaining its cost-effectiveness. When the leakage site is constant, and only the winding mode is altered, it is observed that when the ambient temperature exceeds the liquid temperature in the pipeline, the temperature of the escaping liquid impacts the temperature-measuring fiber due to gravity, registering approximately 2°C higher than the temperature measured directly at the leakage site. The temperature anomaly from field diffusion is significantly less than that caused by the water flow from the leakage impacting the fiber due to gravity. Conversely, when the ambient temperature is lower than the pipeline’s liquid temperature, the opposite occurs. These research findings offer a novel approach for distributed detection in water supply and drainage pipeline leakage.

Recent destructions of structures due to insufficient isolator deformation capacity have led to demands for greater seismic redundancy in seismic isolation design. For a friction pendulum system (FPS), the effect of bidirectional behavior of earthquakes on the maximum response and its effect on friction heating, temperature, and in turn on the maximum response can be significant. However, the extent of these effects under different FPS design parameters and different types of ground motions (GMs) is still not clear. In this study, an analytical model of double concave FPS considering the coupling effect of friction heating and bidirectional behavior was proposed and validated by bidirectional earthquake response orbits, which reflect the characteristics of both GMs and FPSs. Then, the effects of bidirectional GM and corresponding bidirectional temperature change on the response were investigated under different types of strong GMs. Finally, a performance-based design method with a bidirectional-effect-compensation mechanism was proposed. For double concave friction pendulum bearings with PTFE-related layers, it was found that the bidirectional behavior of earthquakes will amplify the maximum isolator displacement by an average of 110–210% (60 MPa) and the maximum superstructure acceleration by an average of 100–140% (60 MPa) under strong GMs (PGV-C1 > 0.2 m/s) and optimum design parameters. The amplification ratio is not only influenced by GM characteristics but also highly related to the design parameters and friction-heating effect of DCFPS.

For vision-based structural displacement measurement and health monitoring, the scale factor (SF) calculation plays a pivotal role in converting the pixel displacement into the actual one. On the other hand, for the current SF calculation methods, the object distance, between the to-be-measured points on the object plane and the optical center of the shooting instrument, has to be measured in advance due to the existence of the pitch and horizontal angle under off-axis measurement. Unfortunately, it is usually inefficient, difficult, and even impossible to obtain the object distances for all the to-be-measured points, especially for the full-field displacement monitoring on huge scale structures. In this paper, a novel SF calculation method was proposed to calculate the full-field scale factor of the host structure under off-axis measurement by combining with the similarity relation of the camera imaging model. With the help of this method, all the scale factors of the to-be-measured points can be calculated accurately and highly efficiently, as long as the object distance of any one point or the geometric dimensions of any feature object on the object surface is provided. In addition, to quantitatively assess the SF calculation accuracy of the method, a static vision-based measurement investigation was firstly conducted, and then, two experimental investigations about the multipoint and full-field structural displacement measurement on a cable-stayed bridge model and a simply supported beam model were conducted to validate the effectiveness and feasibility of the proposed method. Finally, all the results demonstrated that the proposed method exhibits an excellent performance on the SF calculation under off-axis measurement and provides a great potential to be utilized for the vision-based structural multipoint and full-field displacement measurement and health monitoring.

The presence of missing values is common in real-world datasets, so modeling and uncertainty quantification (UQ) of incomplete datasets have gained increasing attention in various research areas, including structural health monitoring (SHM). However, modeling and UQ utilizing incomplete datasets are nontrivial tasks. On the other hand, prediction based on a set of incomplete measured input variables is also an important task, but most existing methods, which are discriminative models, do not possess this capability. Aiming to tackle these two challenges, we propose the two-stage analytical Bayesian copula-based uncertainty quantification (A-BASIC-UQ) using incomplete SHM data. In the modeling stage, the copula-based multivariate joint probability density function (PDF) is modeled directly according to an incomplete dataset without imputation or disposal of any data points. For the univariate marginal PDF, using the measured (nonmissing) values of the corresponding random variable (RV), Bayesian model class selection is conducted to select the most suitable model class. For the Gaussian copula PDF, using the bivariate complete data points of entry-by-entry pairwise data, the optimal parameter vector is obtained from the estimation of the Pearson correlation coefficient. In the prediction stage, the analytical expressions of the predictive PDF, the predicted value and the credible region of the output variables are derived according to a set of incomplete measured input variables. The analytical expression of the predictive PDF is obtained based on the analytical operations on the auxiliary RVs and that of the predicted value and the credible region are obtained based on the analysis of multivariate Gaussian distribution. Therefore, the proposed method does not require numerical integration nor Monte Carlo simulation and does not suffer from computational burden even when there are many variables (say 4 or above). Examples using simulated data and real SHM data are presented to illustrate the capability of the proposed A-BASIC-UQ.