Search Results (28)

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 study systematically investigates the effect of steel fiber content on the fractal evolution characteristics of bending cracks in alkali-activated slag concrete (AASC) beams. A four-point bending test on simply supported beams, combined with digital image correlation (DIC) technology, was employed to quantitatively analyze the fractal dimension of crack propagation paths in AASC beams with steel fiber contents ranging from 0% to 1.4%, using the box-counting method. The relationship between fracture energy and fractal dimension was examined, along with the fractal control mechanisms of mid-span deflection, crack width, and the fractal evolution of fracture toughness parameters. The results revealed that as the steel fiber content increased, the crack fractal dimension decreased from 1.287 to 1.155, while the critical fracture energy of AASC beams increased by approximately 75%. Both mid-span deflection and maximum crack width were positively correlated with the crack fractal dimension, whereas the fractal dimension showed a negative correlation with critical cracking stress and fracture toughness and a positive correlation with the energy release rate. When the steel fiber content exceeded 1.2%, the performance gains began to diminish due to fiber agglomeration effects. Overall, the findings suggest that an optimal steel fiber content range of 1.0% to 1.2% provides the best crack control and mechanical performance, offering a theoretical basis for the design of AASC structures. Full article

This study investigated the fracture behavior of concrete beams with recycled coarse aggregate (RCA) and recycled fine aggregate (RFA) using the box-counting method to measure crack fractal dimensions under load. Beams with RCA showed higher fractal dimensions due to RCA’s lower elastic moduli and compressive strengths, resulting in reduced deformation resistance, ductility, and more late-stage crack propagation. A direct proportional relationship existed between RCA/RFA replacement ratios and crack fractal dimensions. Second-order and third-order polynomial trend surface-fitting techniques were applied to examine the complex relationships among RFA/RCA dosage, applied load, and crack fractal dimension. The results indicated that the RFA dosage had a negative quadratic influence, while load had a positive linear effect, with dosage impact increasing with load. A second-order functional relationship was found between mid-span deflection and crack fractal dimension, reflecting nonlinear behavior consistent with concrete mechanics. This study enhances the understanding of recycled aggregate concrete beam fracture behavior, with the crack fractal dimension serving as a valuable quantitative indicator for damage state and crack complexity assessment. These findings are crucial for engineering design and application, enabling better evaluation of structural performance under various conditions. Full article

The implementation of lean drilling and completion design techniques is a pivotal strategy for the petroleum and natural gas industry to achieve green, low-carbon, and intelligent transformation and innovation. These techniques significantly enhance oil and gas recovery rates. In shale gas development, the shale brittleness index plays a crucial role in evaluating fracturing ability during hydraulic fracturing. Indoor experiments on Gulong shale oil were conducted under a confining pressure of 30 MPa. Based on Rickman’s brittleness evaluation method, this study performed numerical simulations of triaxial compression tests on shale using the finite discrete element method. The fractal dimensions of the fractures formed during shale fragmentation were calculated using the box-counting method. Utilizing the obtained data, a multiple linear regression equation was established with elastic modulus and Poisson’s ratio as the primary variables, and the coefficients were normalized to propose a new brittleness evaluation method. The research findings indicate that the finite discrete element method can effectively simulate the rock fragmentation process, and the established multiple linear regression equation demonstrates high reliability. The weights reassigned for brittleness evaluation based on Rickman’s method are as follows: the coefficient for elastic modulus is 0.43, and the coefficient for Poisson’s ratio is 0.57. Furthermore, the new brittleness evaluation method exhibits a stronger correlation with the brittleness mineral index. The fractal characteristics of crack networks and the relationship between symmetry response and mechanical parameters offer a new theoretical foundation for brittle weight distribution. Additionally, the scale symmetry characteristics inherent in fractal dimensions can serve as a significant indicator for assessing complex crack morphology. Full article

Orthotropic steel box girders and steel bridge decks are commonly applied to bridges. Because of the coupling of original defects and alternating forces, fatigue cracks are likely to appear in the structures. In order to ensure the life span of bridges, methods for automatic crack identification are needed. In this paper, we present a novel approach for crack detection and bridge condition monitoring by integrating convolutional neural networks (CNNs) with graph attention networks (GATs). At first, the original large-sized images are divided into small-sized patches, and these patches are input into a CNN architecture to extract features by decreasing dimensions. Then, the output features of the CNN model are considered as nodes of the graph. Considering the spatial relationship among the patches in the original image, the node from the central patch is connected to the nodes from its neighboring patches to constitute a graph structure, which can be input into a GAT model to learn the relationship among the nodes and update the features. Finally, the output features of GAT can judge whether the central patch contains cracks. Forty original large-sized images are cropped into abundant patches for the training of the CNN-GAT model. With the use of a sliding window technique, the trained CNN-GAT model is capable of finding the patches containing cracks in the test images with large sizes. From the test results, the location and the size of the cracks are exhibited, which indicates that the proposed approach is effective for crack identification in bridge structures. Full article

In response to the challenges of the low accuracy and high misdetection and omission rate of disease detection on feeder roads, an improved Rural-YOLO (SAConv-C2f+C2PSA_CAA+MCSAttention+WIOU) disease detection algorithm is proposed in this paper, which is an enhanced target detection framework based on the YOLOv11 architecture, for the identification of common diseases in the complex feeder road environment. The proposed methodology introduces four key innovations: (1) Switchable Atrous Convolution (SAConv) is introduced into the backbone network to enhance multiscale disease feature extraction under occlusion conditions; (2) Multi-Channel and Spatial Attention (MCSAttention) is constructed in the feature fusion process, and the weight distribution of multiscale diseases is adjusted through adaptive weight redistribution. By adjusting the weight distribution, the model’s sensitivity to subtle disease features is improved. To enhance its ability to discriminate between different disease types, Cross Stage Partial with Parallel Spatial Attention and Channel Adaptive Aggregation (C2PSA_CAA) is constructed at the end of the backbone network. (3) To mitigate category imbalance issues, Weighted Intersection over Union loss (WIoU_loss) is introduced, which helps optimize the bounding box regression process in disease detection and improve the detection of relevant diseases. Based on experimental validation, Rural-YOLO demonstrated superior performance with minimal computational overhead. Only 0.7 M additional parameters is required, and an 8.4% improvement in recall and a 7.8% increase in mAP₅₀ were achieved compared to the initial models. The optimized architecture also reduced the model size by 21%. The test results showed that the proposed model achieved 3.28 M parameters with a computational complexity of 5.0 GFLOPs, meeting the requirements for lightweight deployment scenarios. Cross-validation on multi-scenario public datasets was carried out, and the model’s robustness across diverse road conditions. In the quantitative experiments, the center skeleton method and the maximum internal tangent circle method were used to calculate crack width, and the pixel occupancy ratio method was used to assess the area damage degree of potholes and other diseases. The measurements were converted to actual physical dimensions using a calibrated scale of 0.081:1. Full article

The one-time completion blasting technology for blind shafts is widely used in underground mining, for safety reasons. Efficient blind shaft excavation relies on reasonable cutting blasting technology. To optimize blasting parameters, the impact of explosion stress waves and gases on rock fragmentation is quantitatively analyzed using explosion stress wave theory. A calculation model for the radius R₁ of the crushed zone and the radius R₂ of the fractured zone in rock under the combined action of borehole cutting stress waves and blasting gases is derived and established. Combined with practical engineering examples and the determination method of compensation coefficient C_f, three types of linear cutting patterns, namely six-hole bucket cutting, seven-hole bucket cutting, and nine-hole bucket cutting, are designed. The post-blasting cavity volume and crack length of these three different cutting methods are calculated and analyzed using numerical simulation. Quantitative description of the distribution pattern of blasting-induced cracks in the simulation results of three cutting methods using the box-counting fractal dimension method are presented. Based on this analysis, the nine-hole bucket cutting is selected as the optimal scheme and validated through field application of cutting blasting. The results indicate that the nine-hole bucket cutting blasting scheme for one-time completion of blind shafts, with a designed hole depth of 8 m and a blasthole utilization rate of 93.7%, is an efficient and reasonable technical solution. Full article

The objective of this paper is to develop assessment models to quantitatively evaluate the seismic damage caused to resilient concrete columns intended for buildings located in strong-earthquake-prone regions such as Japan and China. The proposed damage assessment models are based on the fractal analysis of crack patterns on the surface of damaged concrete columns and expressed in the form of a fractal dimension (FD) versus transient drift ratio relationship. To calibrate the proposed damage assessment models, a total of eighty images of crack patterns for eight concrete columns were utilized. All the columns were reinforced by weakly bonded ultra-high-strength (WBUHS) rebars and tested under reversed cyclic loading. The experimental variables covered the shear span ratio of the column, the concrete strength, the axial load ratio, and the amount of steel in the WBUHS rebars. A box-counting algorithm was adopted to calculate or derive the FD of the crack pattern corresponding to each transient drift ratio. The test results reveal that the FD is an efficient image-based quantitative indicator of seismic damage degree for resilient concrete columns and correlates strongly with the transient drift ratio and is subjected to the influence of the shear span ratio. The influence of the other experimental variables on the derived FDs is, if any, little. Based on the test results, a linear equation was developed to define the relationships between the FD and transient drift ratio, and a multi-linear equation was formulated to relate the transient drift ratio to the residual drift ratio, an important index adopted in current design guidelines to measure the repairability of damaged concrete structures. To further verify the efficiency of the drift ratio-based FD in seismic damage assessment, the correlation between the FD and relative stiffness loss (RSL), an indicator used to measure the overall damage degree of concrete structures, was also examined. The driven FD exhibited very strong correlation with RSL, and an empirical equation was developed to reliably assess the overall seismic damage degree of resilient concrete columns with an FD. Full article

This study investigates the effect of thermal cracking on the tensile strength of granite through a combination of experimental testing and numerical simulations. The primary objective is to understand how thermal stress, induced by heat treatment at various temperatures (25 °C to 600 °C), influences crack initiation, propagation, and tensile strength changes. The granite specimens were subjected to Brazilian splitting tests after heat treatment, and the load–displacement curves and tensile strength variations with heat treatment temperature were analyzed. A grain-based model (GBM) was developed to simulate the complex cracking behavior, incorporating the mineral compositions and thermal expansion properties of the granite. The fractal dimension of the cracks was quantified using the box-counting method, and the relationship between fractal dimension and tensile strength was discussed. The results show that the GBM can effectively simulate the microcracking behavior and tensile fracture properties of heat-treated granite, accounting for mineral composition and thermal expansion. Thermal cracks are mainly intergranular tensile cracks, which increase in number with higher temperatures, while under mechanical loading failure is primarily due to intragranular tensile cracks. Higher heat treatment temperatures lead to denser crack networks with greater fractal complexity, reducing tensile strength and creating more tortuous crack propagation paths. Full article

In order to quantitatively describe the influence of aggregate distribution on crack development and peak stress of recycled aggregate concrete, a multifractal spectrum theory was proposed to quantitatively characterize aggregate distribution in specimens. A mesomechanical model of reclaimed aggregate concrete mixed with natural aggregate and artificial aggregate was constructed. Numerical simulation tests were conducted on the uniaxial compression mechanical behavior of 25 groups of sample models with the same proportion and different aggregate distribution forms. Based on the box dimension theory, the multiple fractal spectrum method was used to quantitatively characterize the aggregate distribution form, and the key factors affecting cracks were explored based on the gray correlation degree. The research results show that the aggregate distribution in recycled aggregate concrete has multifractal characteristics. The multifractal spectrum was used to effectively characterize the aggregate distribution pattern, which can enlarge local details and provide new ideas for the quantitative analysis of the damage mode of recycled concrete. Secondly, by establishing a statistical model of the correlation between the multifractal spectrum width of the aggregate distribution pattern and the crack distribution box dimension, it was found that there was a positive correlation between the two, that is, the greater the multifractal spectrum width of the aggregate distribution pattern, the greater the crack box dimension, and the more complex the crack distribution. The complexity of aggregate distribution is closely related to the irregularity and complexity of mesoscopic failure crack propagation in recycled concrete specimens. In addition, gray correlation theory was applied to analyze the key factors affecting the formation of cracks in the specimens. The results showed that aggregate distribution had a first-order correlation with crack formation, and changes in aggregate distribution were an important factor affecting the performance of recycled concrete. Secondly, the poor mechanical properties of NAITZ led to obvious material damage, while NCA and MZ had a significant impact on the skeleton effect in the stress–strain process due to their large areas. This study deepens people’s understanding of the damage characteristics and cracking failure modes of recycled concrete. The study verifies the feasibility of the application of recycled aggregates and provides a valuable reference for engineering practice. Full article

Crack detection is significant to building repair and maintenance; however, conventional inspection is a labor-intensive and time-consuming process for field engineers. This paper proposes a three-step computer vision-based framework to quickly recognize concrete cracks and automatically identify their length, maximum width, and area in damage images. In step one, a region-based convolutional neural network (YOLOv8) is applied to train the crack localizing model. In step two, Gaussian filtering, Canny, and FindContours are integrated to extract the reference contour (a pre-designed seal) to obtain the conversion scale between pixels and millimeter-wise sizes. In step three, the recognized crack bounding box is cropped, and the ApproxPolyDP function and Hough transform are performed to quantify crack dimensions based on the conversion ratio. The developed framework was validated on a dataset of 4630 crack images, and the model training took 150 epochs. Results show that the average crack detection accuracy reaches 95.7%, and the precision of quantified dimensions is over 90%, while the error increases as the crack size grows smaller (increasing to 8% when the crack width is within 1 mm). The proposed method can help engineers to efficiently achieve crack information at building inspection sites, while the reference frame must be pre-marked near the crack, which may limit the scope of application scenarios. In addition, the robustness and accuracy of the developed image processing techniques-based crack quantification algorithm need to be further improved to meet the requirements in real cases when the crack is located within a complex background. Full article

The optimal proportion of alkali-activated steel slag–slag cementitious materials is investigated by considering the combined effects of steel slag content, alkali content, water glass modulus, and water–binder ratio using the Box–Behnken design in response surface methodology. Qualitative and semi-quantitative analyses of X-ray diffraction (XRD) patterns and scanning electron microscopy with energy-dispersive spectroscopy (SEM-EDS) images are conducted. The microstructural mechanism is elucidated based on the chemical composition, surface morphology, and microscale pore (crack) structures of the samples. A microreaction model for the alkali-activated steel slag and slag is proposed. The optimal composition for alkali-activated steel slag–slag cementitious materials is as follows: steel slag content, 38.60%; alkali content, 6.35%; water glass modulus, 1.23; and water–binder ratio, 0.48. The strength values predicted by the response surface model are p_1d = 32.66 MPa, p_7d = 50.46 MPa, and p_28d = 56.87 MPa. XRD analysis confirms that the compressive strength of the sample is not only influenced by the amount of gel formed, but also, to a certain extent, by the CaCO₃ crystals present in the steel slag, which act as nucleation sites. The SEM-EDS results confirm that the gel phase within the system comprises a hydrated calcium silicate gel formed through the reaction of volcanic ash and geopolymer gel formed through geo-polymerization. Analysis of the pore (crack) structure reveals that the compressive strength of the specimens is primarily influenced by porosity, with a secondary influence of the pore fractal dimension. Full article

To better understand the influence of different levels of two-way stress differences on the development of damage in Anchang diametral laminar shale in the northern Qianbei area, a numerical model of laminar shale with a representative fine-scale structure was established by using RFPA3D-CT. A triaxial compression test was conducted; a three-dimensional mesoscale fracture box dimension algorithm based on digital images was generated by using MATLAB R2020b; and the fractal characteristics were quantitatively analyzed. The results showed that under the influence of the horizontal stress ratio and two-way stress, the greater the two-way stress is, the more notable the plastic characteristics of specimen damage are, and the higher the residual strength is. The specimens with lower two-way stress exhibited obvious brittle damage characteristics. The difficulty degree of complex fracture network formation increased with the increase in the horizontal tension ratio, and the degree of increase in the fracture network complexity gradually decreased. At a horizontal stress ratio of 1.25, the fractal dimension was the highest, which indicates that the cracks were the most pronounced. Fracture formation after specimen damage was the most common phenomenon. Under the condition of a lower horizontal stress ratio, a large number of fracture structures could be generated in shale specimens after damage, promoting the expansion of natural fractures. Full article

With the widespread advent of digital technologies, traditional perspectives in rock mechanics research are poised for further expansion. This paper presents a Brazilian test conducted on granite after non-steady-state thermal disturbance at 25 °C, 200 °C, 400 °C, and 600 °C, with detailed documentation of the damage process and failure response using an acoustic emission (AE) apparatus and a digital image correlation (DIC) system. Subsequently, utilizing point cloud data captured by a three-dimensional (3D) laser scanning system, a digital reconstruction of the failed specimen’s fracture surface was accomplished. The 3D fractal characteristics and roughness response of the digitized fracture surface were studied using the box-counting method and least squares approach. Furthermore, texture information of the digitized fracture surface was calculated using the Gray Level Co-occurrence Matrix (GLCM), and statistical characteristics describing the elevation distribution were analyzed. The results elucidate the influence of thermal disturbance temperature on the mechanical parameters of the specimen, acoustic emission behavior, surface strain field evolution, and digital fracture morphology characteristics. The findings indicate a non-linear degradation effect of temperature on the specimen’s tensile strength, with a reduction reaching 80.95% at 600 °C, where acoustic emission activity also peaked. The rising thermal disturbance temperature inhibited the crack initiation load at the specimen’s center but expanded the high-strain concentration areas and the growth rate of horizontal displacement. Additionally, varying degrees of linear or non-linear relationships were discovered between thermal disturbance temperature and the 3D fractal dimension of the fracture surface, average roughness (R_a), peak roughness (R_z), and root mean square roughness (R_q), confirming the potential of Rsm in predicting the 3D fractal dimension of Brazilian test fracture surfaces. The study of the GLCM of the digitized 3D fracture surface demonstrated a high dependency of its four second-order statistical measures on thermal disturbance temperature. Finally, the statistical parameters of the fracture surface’s elevation values showed a significant non-linear relationship with thermal disturbance temperature, with a critical temperature point likely existing between 400 and 600 °C that could precipitate a sudden change in the fracture surface’s elevation characteristics. Full article

The utilisation of box girders with corrugated steel webs (CSWs) represents an innovative approach to bridge superstructure design that has garnered substantial popularity worldwide, with a notable prevalence in both Asia and Europe. Compared with traditional box girders, they avoid web cracking, improving the prestressing efficiency and bridge spanning ability. As an innovative box girder, a corrugated web can increase the cantilever length and transverse stiffness, and at the same time, it reduces the dead weight of the bridge deck. However, little research has been conducted on the mechanical properties of this novel spine-like box girder with CSWs, especially its transverse performance, although it has been used in many applications. In this paper, the effect of the web form on the behaviour of box girders is introduced. Therefore, three representative three-dimensional (3D) finite-element models (i.e., corrugated web box girder, flat web box girder, and ordinary equivalent concrete web box girder) have been established to quantitatively investigate the influence of corrugated web stiffness on transverse stress under the action of gravity and vehicle loads. Generally, significant differences in the mechanical performance of box girders with CSWs have been observed compared with conventional box girders with concrete webs. Additionally, parametric studies to investigate the influences of the corrugation dimensions (in term of the corrugation height, web thickness, panel width, web height and elastic modulus) on the transverse stiffness of such bridges are analyzed. The results show that a new stiffness formula can be put forward to consider the effect of web height, and a high-strength steel web needs to be developed urgently for box girders with CSWs in the near future. Overall, the results of this investigation can be used as a reference for transverse designing and segmental construction of similar projects. Full article