Because of the inconvenience of installing sensors in a buried pipeline, an acoustic emission sensor is initially proposed for collecting and analyzing leakage signals inside the pipeline. Four operating conditions of a fluid-filled pipeline are established and a support vector machine (SVM) method is used to accurately classify the leakage condition of the pipeline. Wavelet decomposition and empirical mode decomposition (EMD) methods are initially used in denoising these signals to address the problem in which original leakage acoustic emission signals contain too much noise. Signals with more information and energy are then reconstructed. The time-delay estimation method is finally used to accurately locate the leakage source in the pipeline. The results show that by using SVM, wavelet decomposition and EMD methods, leakage detection in a liquid-filled pipe with built-in acoustic emission sensors is effective and accurate and provides a reference value for real-time online monitoring of pipeline operational status with broad application prospects.

Damping is a fundamental characteristic of bridge structures, reflecting their ability to dissipate energy during vibration. In the design and maintenance of bridges, the damping ratio has a direct impact on the safety and service life of the structure, thus affecting its sustainability. Currently, there is no suitable theoretical method for estimating structural damping at the design stage. Therefore, the modal damping ratio of a completed or under-construction bridge can only be obtained through field dynamic tests to ensure compliance with design specifications. To summarize the latest research findings on bridge structure damping models and identification methods, and to advance the development of damping identification techniques, this paper provides an in-depth review from several perspectives: Firstly, it offers a comprehensive analysis of the theoretical framework for structural damping. Secondly, it summarizes the damping models proposed by researchers from various countries. Thirdly, it reviews the research progress on identifying the modal damping ratio of bridge structures using time domain, frequency domain, and time-frequency domain methods based on environmental excitation. It also summarizes the methods and current status of identifying the modal damping ratio using artificial excitation. Finally, the future prospects and conclusions are discussed from three aspects: damping theory, test and identification method and data processing. This research and summary provide a solid theoretical foundation for advancing bridge structural damping theory and identification methods and offer valuable references for bridge operation and maintenance, as well as damage identification. From the perspective of modal parameter identification, it provides a theoretical basis for the sustainable development of bridges.

The use of deep learning to interpret ground penetrating radar (GPR) data for automated target detection has become a rapidly developing field. However, the performance of deep learning networks is significantly affected by the quantity and diversity of the available GPR data for training. The finite-difference time-domain (FDTD) method is commonly used to supplement datasets. However, there are considerable differences between FDTD-simulated and real GPR data, which limit the effectiveness of using FDTD data for training. To address these challenges, we propose a generative adversarial network (GAN)-based domain adaptation network for detecting target reflections in GPR data. This network combines the You Only Look Once (YOLO) network with a domain adaptation (DA) module, forming a GAN-based DA-YOLO that aligns features extracted from simulated and real GPR data to enhance the performance of the network on real GPR data. To further improve the network’s performance on real GPR data, a pseudo-labeling strategy is employed. This approach leverages the pseudo-labels generated by the network to iteratively refine and improve its accuracy. To validate the proposed method, we conduct comparative experiments to evaluate the impact of key factors, including network architecture, pseudo-labeling thresholds, and source-target domain discrepancies, on feature alignment and the improvement of detection performance for target domain data. The effectiveness of the proposed method is validated using real-world GPR data collected from a railway tunnel inspection. The results demonstrate that the proposed DA-YOLO network achieves the best detection performance on real data, with recall, precision, F1 score, and mAP values of 81.25%, 86.67%, 83.87%, and 69.30%, respectively, outperforming other networks that do not utilize domain adaptation methods. This study provides a solution for situations where real GPR data is scarce and only FDTD-simulated data can be used as training data.

In recent years, concrete-filled steel tubular (CFST) bridges have been widely used in bridge construction. However, interface disengaging is one of the bridges’ most common defects. It affects not only the confinement of the steel tube to the core concrete, but also the bearing capacity and the structure durability. In order to solve the problem of disengaging at the interface of steel tube-confined concrete and ensure the integrity of the structure, a new method based on a distributed temperature detection system is proposed to identify disengaging at the interface between steel tube and concrete. Through the analysis of experiments and numerical simulation, it was found that when the structure surface was heated, the temperature of the void area was higher than that of the solid area. The temperature of the maximum void height was the highest, the temperature increase rate was the fastest, and the cooling rate was the slowest. The boundary of the void suffered from sudden changes in temperature. The more serious the void, the greater was the temperature difference between the void and the solid. After changing the heating direction, the temperature distribution could still be identified. This study achieved simplicity and efficiency with respect to the detection of interface separation.

The crosshole ground-penetrating radar (GPR) technique is widely used to characterize subsurface structures, yet the interpretation of crosshole GPR data involves solving non-linear and ill-posed inverse problems. In this work, we developed a generative adversarial network (GAN)-based inversion framework to translate crosshole GPR images to their corresponding 2D defect reconstruction images automatically. This approach uses fully connected layers to extract global features from crosshole GPR images and employs a series of cascaded U-Net structures to produce high-resolution defect reconstruction results. The feasibility of the proposed framework was demonstrated on a synthetic crosshole GPR dataset created with the finite-difference time-domain (FDTD) method and real-world data from a field experiment. Our inversion network obtained recognition accuracy of 91.36%, structural similarity index measure (SSIM) of 0.93, and RAscore of 91.77 on the test dataset. Furthermore, comparisons with ray-based tomography and full-waveform inversion (FWI) suggest that the proposed method provides a good balance between inversion accuracy and efficiency and has the best generalization when inverting actual measured crosshole GPR data.

As a new kind of composite structures and its advantages, the using of steel tube confined reinforced concrete (STCRC) columns have received increasing attention in civil engineering. Acoustic emission (AE) technique is applied to monitor the damage process of STCRC columns during an uniaxial compression test. The aim of this study is to investigate the damage evaluation and failure mechanisms of the STCRC columns from the perspective of microscopic damage. Typical AE parameters are extracted to quantify different damage condition and identify the critical point. Peak-frequency analysis classify the damage signal into groups, representing different AE generation mechanisms. Extended AE features, such as the RA value and the average frequency are calculated to discriminate the cracking modes of core concrete. The probability based Gaussian mixture model (GMM) are proposed for unsupervised damage pattern recognition. All presented AE results are good in accordance with the observed experimental outcomes. The AE technology enabled evaluating the damage condition, identifying critical point, disclosing the failure mechanism, and classifying damage modes for steel confined RC columns effectively.