Search Results (160)

To address the inefficiency of traditional concrete crack detection methods and the heavy reliance of supervised learning on extensive labeled data, in this study, an intelligent assessment method of concrete damage based on pseudo-label semi-supervised learning and fractal geometry theory is proposed to solve two core tasks: one is binary classification of pixel-level cracks, and the other is multi-category assessment of damage state based on crack morphology. Using three-channel RGB images as input, a dual-path collaborative training framework based on U-Net encoder–decoder architecture is constructed, and a binary segmentation mask of the same size is output to achieve the accurate segmentation of cracks at the pixel level. By constructing a dual-path collaborative training framework and employing a dynamic pseudo-label refinement mechanism, the model achieves an F1-score of 0.883 using only 50% labeled data—a mere 1.3% decrease compared to the fully supervised benchmark DeepCrack (F1 = 0.896)—while reducing manual annotation costs by over 60%. Furthermore, a quantitative correlation model between crack fractal characteristics and structural damage severity is established by combining a U-Net segmentation network with the differential box-counting algorithm. The experimental results demonstrate that under a cyclic loading of 147.6–221.4 kN, the fractal dimension monotonically increases from 1.073 (moderate damage) to 1.189 (failure), with 100% accuracy in damage state identification, closely aligning with the degradation trend of macroscopic mechanical properties. In complex crack scenarios, the model attains a recall rate (Re = 0.882), surpassing U-Net by 13.9%, with significantly enhanced edge reconstruction precision. Compared with the mainstream models, this method effectively alleviates the problem of data annotation dependence through a semi-supervised strategy while maintaining high accuracy. It provides an efficient structural health monitoring solution for engineering practice, which is of great value to promote the application of intelligent detection technology in infrastructure operation and maintenance. Full article

This paper presents a UAV-based building crack real-time detection system that integrates an improved YOLOv8 algorithm with Linear Active Disturbance Rejection Control (LADRC). The system is equipped with a high-resolution camera and sensors to capture high-definition images and height information. First, a trajectory tracking controller based on LADRC was designed for the UAV, which uses a linear extended state observer to estimate and compensate for unknown disturbances such as wind interference, significantly enhancing the flight stability of the UAV in complex environments and ensuring stable crack image acquisition. Secondly, we integrated Convolutional Block Attention Module (CBAM) into the YOLOv8 model, dynamically enhancing crack feature extraction through both channel and spatial attention mechanisms, thereby improving recognition robustness in complex backgrounds. Lastly, a skeleton extraction algorithm was applied for the secondary processing of the segmented cracks, enabling precise calculations of crack length and average width and outputting the results to a user interface for visualization. The experimental results demonstrate that the system successfully identifies and extracts crack regions, accurately calculates crack dimensions, and enables real-time monitoring through high-speed data transmission to the ground station. Compared to traditional manual inspection methods, the system significantly improves detection efficiency while maintaining high accuracy and reliability. Full article

The identification of structural damage types remains a key challenge in electromechanical impedance/admittance (EMI/EMA)-based structural health monitoring realm. This paper proposed a damage classification approach for concrete structures by using integrating discrete wavelet transform (DWT) decomposition of EMA signatures with supervised machine learning. In this approach, the EMA signals of arranged piezoelectric ceramic (PZT) patches were successively measured at initial undamaged and post-damaged states, and the signals were decomposed and processed using the DWT technique to derive indicators including the wavelet energy, the variance, the mean, and the entropy. Then these indicators, incorporated with traditional ones including root mean square deviation (RMSD), baseline-changeable RMSD named RMSDk, correlation coefficient (CC), and mean absolute percentage deviation (MAPD), were processed by a support vector machine (SVM) model, and finally damage type could be automatically classified and identified. To validate the approach, experiments on a full-scale reinforced concrete (RC) slab and application to a practical tunnel segment RC slab structure instrumented with multiple PZT patches were conducted to classify severe transverse cracking and minor crack/impact damages. Experimental and application results cogently demonstrated that the proposed DWT-based approach can precisely classify different types of damage on concrete structures with higher accuracy than traditional ones, highlighting the potential of the DWT-decomposed EMA signatures for damage characterization in concrete infrastructure. Full article

Cracks and seepage in dam structures pose a serious risk to their safety, yet traditional inspection methods often fall short when it comes to detecting shallow or early-stage fractures. This study proposes a new approach that uses spectral response analysis to quickly identify signs of seepage in concrete dams. Researchers developed a three-layer model—representing the concrete, a seepage zone, and water—to better understand how cracks affect the way electrical signals behave, thereby inverting the state of the dam based on how electrical signals behave in actual engineering measurements. Through computer simulations and lab experiments, the team explored how changes in the resistivity and thickness of the seepage layer, along with the resistivity of surrounding water, influence key indicators like impedance and signal angle. The results show that the “spectrum-based method” can effectively detect seepage in concrete cracks of dams, and the measurement method of the “spectral quadrupole method” based on the “spectrum-based method” is highly sensitive to these variations, making it a promising tool for spotting early seepage. Field tests backed up the lab findings, confirming that this method is significantly better than traditional techniques at detecting cracks less than a meter deep and identifying early signs of water intrusion. It could provide dam inspectors with a more reliable way to monitor structural health and prevent potential failures. Full article

With the extensive application of 3D-printed composites across multiple industries, the investigation into their structural reliability under complex loading conditions has become a critical research focus. This study comprehensively employs acoustic emission (AE) monitoring, digital image correlation (DIC) measurement, and micro-computed tomography (Micro-CT) visualization techniques to explore the progressive damage behavior of 3D-printed sandwich-structured composites reinforced with continuous carbon fiber sheets under three-point bending. Mechanical tests show that increasing the fiber content of face sheets from 10% to 20% enhances average bending strength by 56%, while low fiber content compromises stiffness and load-bearing capacity. AE analysis categorizes damage modes into matrix cracking (<50 kHz), debonding/delamination (50–150 kHz), and fiber breakage (>150 kHz) using k-means clustering algorithms. DIC measurement reveals significant structural deformation processes during damage progression. The AE-DIC-Micro-CT combination demonstrates an initial undamaged state, followed by damage initiation and propagation in the subsequent stages. This integrated approach provides an effective method for damage assessment, guiding the design and reliability improvement of 3D-printed composites. Full article

It is vital to stabilize pillar dams in underground reservoirs in coal mine goafs to protect groundwater resources and quarry safety, practice green mining, and protect the ecological environment. Considering the actual occurrence of coal pillar dams in underground reservoirs, acoustic emission (AE) mechanical tests were performed on dry, naturally absorbed, and soaked coal samples. According to the mechanical analysis, Quantitative analysis revealed that dry samples exhibited the highest mechanical parameters (peak strength: 12.3 ± 0.8 MPa; elastic modulus: 1.45 ± 0.12 GPa), followed by natural absorption (peak strength: 9.7 ± 0.6 MPa; elastic modulus: 1.02 ± 0.09 GPa), and soaked absorption showed the lowest values (peak strength: 7.2 ± 0.5 MPa; elastic modulus: 0.78 ± 0.07 GPa). The rate of mechanical deterioration increased by ~25% per 1% increase in moisture content. It was identified that the internal crack development presented a macrofracture surface initiating at the sample center and expanding radially outward, and gradually expanding to the edges by adopting AE seismic source localization and the K-means clustering algorithm. Soaked absorption was easier to produce shear cracks than natural absorption, and a higher water content increased the likelihood. The b-value of the AE damage evaluation index based on crack development was negatively correlated with the rock damage state, and the S-value was positively correlated, and both effectively characterized it. The research results can offer reference and guidance for the support design, monitoring, and warning of coal pillar dams in underground reservoirs. (The samples were tested under two moisture conditions: (1) ‘Soaked absorption’—samples fully saturated by immersion in water for 24 h, and (2) ‘Natural absorption’—samples equilibrated at 50% relative humidity and 25 °C for 7 days). Full article

Part defects in additive manufacturing (AM) operations like laser sintering (LS) can negatively affect the quality and integrity of the manufactured parts. Therefore, it is important to understand and mitigate these part defects to improve the performance and safety of the manufactured parts. Integrating machine learning to detect part defects in AM can enable efficient, fast, and automated real-time monitoring, reducing the need for labor-intensive manual inspections. In this work, a novel approach incorporating a lightweight Visual Geometry Group (VGG) structure with soft attention is presented to detect powder bed defects (such as cracks, powder bed ditches, etc.) in laser sintering processes. The model was evaluated on a publicly accessible dataset (called LS Powder bed defects) containing 8514 images of powder bed images pre-split into training, validation, and testing sets. The proposed methodology achieved an accuracy of 98.40%, a precision of 97.45%, a recall of 99.40%, and an f1-score of 98.42% with a computation complexity of 0.797 GMACs. Furthermore, the proposed method achieved better performance than the state-of-the-art in terms of accuracy, precision, recall, and f1-score on LS powder bed images, while requiring lower computational power for real-time application. Full article

Surface cracks are key indicators of landslide deformation, crucial for early landslide identification and deformation pattern analysis. However, due to the complex terrain and landslide extent, manual surveys or traditional digital image processing often face challenges with efficiency, precision, and interference susceptibility in detecting these cracks. Therefore, this study proposes a comprehensive automated pipeline to enhance the efficiency and accuracy of landslide surface crack detection. First, high-resolution images of landslide areas are collected using unmanned aerial vehicles (UAVs) to generate a digital orthophoto map (DOM). Subsequently, building upon the U-Net architecture, an improved encoder–decoder semantic segmentation network (IEDSSNet) was proposed to segment surface cracks from the images with complex backgrounds. The model enhances the extraction of crack features by integrating residual blocks and attention mechanisms within the encoder. Additionally, it incorporates multi-scale skip connections and channel-wise cross attention modules in the decoder to improve feature reconstruction capabilities. Finally, post-processing techniques such as morphological operations and dimension measurements were applied to crack masks to generate crack inventories. The proposed method was validated using data from the Heifangtai loess landslide in Gansu Province. Results demonstrate its superiority over current state-of-the-art semantic segmentation networks and open-source crack detection networks, achieving F1 scores and IOU of 82.11% and 69.65%, respectively—representing improvements of 3.31% and 4.63% over the baseline U-Net model. Furthermore, it maintained optimal performance with demonstrated generalization capability under varying illumination conditions. In this area, a total of 1658 surface cracks were detected and cataloged, achieving an accuracy of 85.22%. The method proposed in this study demonstrates strong performance in detecting surface cracks in landslide areas, providing essential data for landslide monitoring, early warning systems, and mitigation strategies. Full article

To investigate the degradation mechanisms of the surrounding rock in abandoned mine roadways used for oil storage, this study combined uniaxial compression tests with digital image correlation (DIC), scanning electron microscopy (SEM), and other techniques to analyze the evolution of the rock mechanical properties under the coupled effects of oil–water soaking and initial damage. The results indicate that oil–water soaking induces the loss of silicon elements and the deterioration of microstructure, leading to surface peeling, crack propagation, and increased porosity of the sample. The compressive strength decreases linearly with the soaking time. Acoustic emission (AE) monitoring showed that after 24 h of soaking, the maximum ringing count rate and cumulative count decreased by 81.7% and 80.4%, respectively, compared to the dry state. As the liquid saturation increases, the failure mode transitions from tension dominated to shear failure. The synergistic effect of initial damage and oil–water erosion weakens the rock’s energy storage capacity, with the energy storage limit decreasing by 45.6%, leading to reduced resistance to external forces. Full article

Crack segmentation is crucial for health monitoring and preventive maintenance of concrete structures. However, the complex morphologies of cracks and the limited resources of mobile devices pose challenges for accurate and efficient segmentation. To address this, we propose an efficient dynamic-state-space-enhanced network termed DSS-MobileNetV3 for crack segmentation. The DSS-MobileNetV3 adopts a U-shaped encoder–decoder architecture, and a dynamic-state-space (DSS) block is designed into the encoder to improve the MobileNetV3 bottleneck module in modeling global dependencies. The DSS block improves the MobileNetV3 model in structural perception and global dependency modeling for complex crack morphologies by integrating dynamic snake convolution and a state space model. The decoder utilizes the upsampling and depthwise separable convolution to progressively decode and efficiently restore the spatial resolution. In addition, to suppress complex noise in the image background and highlight crack textures, the strip pooling module is introduced into the skip connection between the encoder and decoder for performance enhancement. Extensive experiments are conducted on three public crack datasets, and the proposed DSS-MobileNetV3 achieves SOTA performance in both accuracy and efficiency. Full article

Convolutional Neural Networks (CNNs) have been demonstrated to be one of the most powerful methods for image recognition, being applied in many fields, including civil and structural health monitoring in infrastructure asset management. Current State-of-the-Art CNN models are now accessible as open-source and available on several Artificial Intelligence (AI) platforms, with TensorFlow being widely used. Besides CNN models, Vision Transformers (ViTs) have recently emerged as a competitive alternative. Several demonstrations have indicated that ViT models, in many instances, outperform the current CNNs by almost four times in terms of computational efficiency and accuracy. This paper presents an investigation into defect detection for civil and structural components using CNN and ViT models available on TensorFlow. An empirical study was conducted using a database of cracks. The severity of crack is categorized into binary states: “with crack” and “without crack”. The results confirm that the accuracies of both CNN and ViT models exceed 95% after 100 epochs of training, with no significant difference observed between them for binary classification. Notably, the cost of this AI-based approach with images taken by lightweight and low-cost drones is considerably lower compared to high-speed inspection cars, while still delivering an expected level of predictive accuracy. Full article

The welded joints in steel bridges have a complicated structure, and their fatigue life is mainly determined by the real stress under the coupling effect of vehicle load stress, as well as weld-induced residual stress and its relaxation. Traditional fatigue analysis methods are inadequate for effectively accounting for weld-induced residual stress and its relaxation, resulting in a significant discrepancy between the predicted fatigue life and the actual fatigue cracking time. A fatigue damage assessment model of welded joints was developed in this study, considering weld-induced residual stress and its relaxation under vehicle load stress. A multi-scale finite element model (FEM) for vehicle-induced coupled analysis was established to investigate the weld-induced initial residual stress and its relaxation effect associated with cyclic bend fatigue due to vehicles. The fatigue damage assessment, considering the welding residual stress and its relaxation, was performed based on the S–N curve model from metal fatigue theory and Miner’s linear damage theory. Based on this, the impact of variations in traffic load on fatigue life was forecasted. The results show that (1) the state of tension or compression in vehicle load stress notably impacts the residual stress relaxation effect observed in welded joints, of which the relaxation magnitude of the von Mises stress amounts to 81.2% of the average vehicle load stress value under tensile stress working conditions; (2) the predicted life of deck-to-rib welded joints is 28.26 years, based on traffic data from Jiangyin Bridge, which is closer to the monitored fatigue cracking life when compared with the Eurocode 3 and AASHTO LRFD standards; and (3) when vehicle weight and traffic volume increase by 30%, the fatigue life significantly drops to just 9.25 and 12.13 years, receptively. Full article

This study established a dynamic impact simulation system for a coral limestone cement composite subjected to bidirectional stress confinement conditions by using a coupled method of continuous medium FDM (a coupled continuum-discontinuum approach integrating finite difference continuum modeling (FDM) and the discrete element method (DEM) granular analysis), and verified its accuracy through indoor experiments. The study first conducted dynamic mechanical performance tests on reef limestone concrete using an SHPB experimental device, exploring the effects of the strain-rate governed high-rate response, energy evolution, and failure modes. Subsequently, an FDM-DEM coupled model was used to simulate the impact-induced behavior of concrete at multiaxial stress conditions and constraint conditions, analyzing the strain-rate dependent performance of concrete exposed to biaxial monotonic loading. Test outcomes indicate that the increase in strain rate significantly enhanced the dynamic peak stress, and the collapse behavior shifted from type I to type II. As static loading in the σ₂ direction increased, the dynamic peak stress in the σ₁ direction decreased, while the dynamic peak stress in the σ₂ direction increased, indicating that the constraint stress in the σ₂ direction had an inhibitory effect on the sample’s failure. Through the time-history monitoring and analysis of cracks, it was found that the internal crack growth rate accelerated as the stress increased, while the crack growth tended to stabilize when the stress decreased. Additionally, this study explored the effect of stress constraints on the fragmentation patterns, revealing changes in the failure modes and crack distributions of the sample under different stress states, providing a theoretical basis and technical support for island and reef construction and engineering protection. Full article

The growing application of Fiber-Reinforced Polymer (FRP) composites in rehabilitating deteriorating concrete infrastructure underscores the need for reliable, cost-effective, and automated nondestructive testing (NDT) methods. This review provides a comprehensive analysis of existing and emerging NDT techniques used to assess externally bonded FRP (EB-FRP) systems, emphasizing their accuracy, limitations, and practicality. Various NDT methods, including Ground-Penetrating Radar (GPR), Phased Array Ultrasonic Testing (PAUT), Infrared Thermography (IRT), Acoustic Emission (AE), and Impact–Echo (IE), are critically evaluated in terms of their effectiveness in detecting debonding, voids, delaminations, and other defects. Recent technological advancements, particularly the integration of artificial intelligence (AI) and machine learning (ML) in NDT applications, have significantly improved defect characterization, automated inspections, and real-time data analysis. This review highlights AI-driven NDT approaches such as automated crack detection, hybrid NDT frameworks, and drone-assisted thermographic inspections, which enhance accuracy and efficiency in large-scale infrastructure assessments. Additionally, economic considerations and cost–performance trade-offs are analyzed, addressing the feasibility of different NDT methods in real-world FRP-strengthened structures. Finally, the review identifies key research gaps, including the need for standardization in FRP-NDT applications, AI-enhanced defect quantification, and hybrid inspection techniques. By consolidating state-of-the-art research and emerging innovations, this paper serves as a valuable resource for engineers, researchers, and practitioners involved in the assessment, monitoring, and maintenance of FRP-strengthened concrete structures. Full article

Granite is widely used in laboratory rockburst simulations due to its exceptional strength, brittleness, and uniform composition. This study employs a true triaxial loading system to replicate asymmetric stress states near free surfaces, allowing precise control of three-dimensional stresses to simulate strain-mode rockbursts. Advanced monitoring tools, such as acoustic emission (AE) and high-speed imaging, were used to investigate the evolution process, failure mechanisms, and monitoring strategies. The evolution of strain-mode rockbursts is divided into five stages: stress accumulation, crack initiation, critical instability, rockburst occurrence, and residual stress adjustment. Each stage exhibits dynamic responses and progressive energy release. Failure is governed by a tension–shear coexistence mechanism, where vertical splitting and diagonal shear fractures near free surfaces lead to V-shaped craters and violent rock fragment ejection. This reflects the brittle nature of granite under high-stress conditions. The AE monitoring proved highly effective in identifying rockburst precursors, with key indicators including quiet periods of low AE activity and sudden surges in AE counts, coupled with ‘V-shaped’ b-value troughs, offering reliable early warning signals. These findings provide critical insights into strain-mode rockburst dynamics, highlighting the transition from elastic deformation to dynamic failure and the role of energy release mechanisms. Full article