Search Results (116)

Efficient and accurate recognition of highway pavement cracks is crucial for the timely maintenance and long-term use of expressways. Among the existing crack acquisition methods, human-based approaches are inefficient, whereas carrier-based automated methods are expensive. Additionally, both methods present challenges related to traffic obstruction and safety risks. To address these challenges, we propose a fixed pan-tilt-zoom (PTZ) vision-based highway pavement crack recognition workflow. Pavement cracks often exhibit complex textures with blurred boundaries, low contrast, and discontinuous pixels, leading to missed and false detection. To mitigate these issues, we introduce an algorithm named contrast-enhanced feature reconstruction (CEFR), which consists of three parts: comparison-based pixel transformation, nonlinear stretching, and generating a saliency map. CEFR is an image pre-processing algorithm that enhances crack edges and establishes uniform inner-crack characteristics, thereby increasing the contrast between cracks and the background. Extensive experiments demonstrate that CEFR improves recognition performance, yielding increases of 3.1% in F1-score, 2.6% in mAP@0.5, and 4.6% in mAP@0.5:0.95, compared with the dataset without CEFR. The effectiveness and generalisability of CEFR are validated across multiple models, datasets, and tasks, confirming its applicability for highway maintenance engineering. Full article

Accurate detection of slope cracks plays a crucial role in early landslide disaster warning; however, traditional approaches often struggle to identify fine and irregular cracks. This study introduces a novel deep learning model, Crack-Net, which leverages a multi-modal feature fusion mechanism and is developed using transfer learning. To resolve the blurred representation of small-scale cracks, a nonlinear frequency-domain mapping module is employed to decouple amplitude and phase information, while a cross-domain attention mechanism facilitates adaptive feature fusion. In addition, a deep feature fusion module integrating deformable convolution and a dual attention mechanism is embedded within the encoder–decoder architecture to enhance multi-scale feature interactions and preserve crack topology. The model is pre-trained on the CrackVision12K dataset and fine-tuned on a custom dataset of slope cracks, effectively addressing performance degradation in small-sample scenarios. Experimental results show that Crack-Net achieves an average accuracy of 92.1%, outperforming existing models such as DeepLabV3 and CrackFormer by 9.4% and 5.4%, respectively. Furthermore, the use of transfer learning improves the average precision by 1.6%, highlighting the model’s strong generalization capability and practical effectiveness in real-world slope crack detection. Full article

Different from general underground engineering, the micro-damage prior to failure of the surrounding rock has a significant influence on the geological disposal of high-level radioactive waste. However, the quantitative research on pre-failure dilatancy damage characteristics and stress path influence of hard brittle rocks under high stress levels is insufficient currently, and especially, the stress path under simultaneous unloading of axial and confining pressures is rarely discussed. Therefore, three representative mechanical experimental studies were conducted on the Beishan granite in the pre-selected area for high-level radioactive waste (HLW) geological disposal in China, including increasing axial pressure with constant confining pressure (path I), increasing axial pressure with unloading confining pressure (path II), and simultaneous unloading of axial and confining pressures (path III). Using the deviatoric stress ratio as a reference, the evolution laws and characteristics of stress–strain relationships, deformation modulus, generalized Poisson’s ratio, dilatancy index, and dilation angle during the path bifurcation stage were quantitatively analyzed and compared. The results indicate that macro-deformation and the plastic dilatancy process exhibit strong path dependency. The critical value and growth gradient of the dilatancy parameter for path I are both the smallest, and the suppressive effect of the initial confining pressure is the most significant. The dilation gradient of path II is the largest, but the degree of dilatancy before the critical point is the smallest due to its susceptibility to fracture. The critical values of the dilatancy parameters for path III are the highest and are minimally affected by the initial confining pressure, indicating the most significant dilatancy properties. Establish the relationship between the deformation parameters and the crack-induced volumetric strain and define the damage variable accordingly. The critical damage state and the damage accumulation process under various stress paths were examined in detail. The results show that the damage evolution is obviously differentiated with the bifurcation of the stress paths, and three different types of damage curve clusters are formed, indicating that the damage accumulation path is highly dependent on the stress path. The research findings quantitatively reveal the differences in deformation response and damage characteristics of Beishan granite under varying stress paths, providing a foundation for studying the nonlinear mechanical behavior and damage failure mechanisms of hard brittle rock under complex loading conditions. Full article

Although oily sludge is an industrial waste and difficult to separate, its calorific value can still reach 6000 cal/g, thus possessing significant recycling value. This study compares various types of non-thermal plasma for refining oily sludge. The pre-treatment technology utilized filtration combined with solvent extraction to extract the oil portion from the oily sludge. Subsequently, two types of non-thermal plasma, DC streamer discharge and dielectric plasma discharge, were used to crack and activate the oily sludge under different operating conditions. The fuel compositions and properties of the refined fuel treated by two types of non-thermal plasma were compared. The elemental carbon and oxygen of the oily sludge after treatment in a direct DBD plasma reactor for 8 min were 1.96 wt.% less and 1.38 wt.% higher than those of commercial diesel. The research results indicate that the pre-treatment process can effectively improve the refined fuel properties. After pre-treatment, the calorific value of the primary product from the oily sludge can reach 10,598 cal/g. However, the carbon residue of the oily sludge after pre-treatment remained as high as 5.58 wt.%, which implied that further refining processes are required. The streamer discharge plasma reactor used a tungsten needle tip as a high-voltage electrode, leading to a rather small treated range. Corona discharge and arc formation are prone to being produced during the plasma action. Moreover, the addition of quartz glass beads can form a protruding area on the surface of the oily sludge, generating an increase in the reacting surface of the oily sludge, and hence an enhancement of treatment efficiency, in turn. The direct treatment of DBD plasma can thus have a wider and more uniform operating range of plasma generation and a superior efficiency of plasma reaction. Therefore, a direct DBD type of non-thermal equilibrium plasma reactor is preferable to treat oily sludge among those three types of plasma reactor designs. Additionally, when the plasma voltage is increased, it effectively enhances fuel properties. Full article

The Dolomite reservoir of the Maokou Formation is rich in gas resources in the central Sichuan Basin. Acid fracturing is an important technical means to improve reservoir permeability and productivity. The interaction mode of the dolomite and limestone acid system will affect the effect of reservoir reconstruction. In order to clarify the influence of complex structure on fracture morphology, we explore the fracturing effect of different acid systems. Physical simulation experiments of true triaxial acid fracturing were carried out with two acid systems and downhole full-diameter cores. The experimental results show: (1) After the carbonate rock is subjected to acid fracturing using a “self-generated acid + gel acid” system, the fracture pressure drops significantly by up to 60%. The morphology of the acid-eroded fractures becomes more complex, with an increase in geometric complexity of about 28% compared to a single acid solution system. It is prone to form three-dimensional “spoon” shaped fractures, and the surface of the acid-eroded fractures shows light yellow acid erosion marks. Analysis of the acid erosion marks indicates that the erosion depth on the fracture surface reaches 0.8–1.2 mm, which is deeper than the 0.2 mm erosion depth achieved with a single system. (2) Acid solution is difficult to penetrate randomly distributed calcite veins with a low porosity and permeability structure. When the fracture meets the calcite vein, the penetration rate of acid solution drops sharply to 15–20% of the initial value, resulting in a reduction of about 62% of the acid erosion area in the limestone section behind. And the acid erosion traces in the limestone behind the calcite vein are significantly reduced. The acid erosion cracks are easy to open on the weak surface between dolomite and limestone, causing the fracture to turn. (3) The results of field engineering and experiment are consistent, and injecting authigenic acid first in the process of reservoir reconstruction is helpful to remove pollution. The recovery rate of near-well permeability is more than 85% with pre-generated acid. Reinjection of gelled acid can effectively communicate the natural weak surface and increase the complexity of cracks. The average daily oil production of the completed well was increased from 7.8 m³ to 22.5 m³, and the increase factor reached 2.88. Full article

To address roof overhang hazards (e.g., rock bursts and gas accumulation) in high-gas coal mines, this study proposes a static stress intervention method for controlled directional roof collapse. Using the 150110 fully mechanized face at Yiyuan Coal Mine as a case study, we investigate the mechanical mechanism of static stress intervention-induced roof collapse through theoretical modeling and FLAC3D simulations in the absence of pre-cracks. The study reveals that advanced boreholes filled with static expansion agents generate stress concentration zones along the drilling array. When superimposed with mining-induced stresses, this configuration induces tensile failure preferentially at borehole locations, thereby achieving controlled directional roof collapse. Theoretical calculations indicate that roof fracturing occurs at predetermined locations when expansion pressure reaches ≥9.11 MPa. FLAC3D simulations analyzed stress redistribution and plastic zone evolution under combined static and mining-induced stresses, demonstrating the method’s efficacy in optimizing roadway stability. Field trials implement spaced boreholes (65 mm diameter, 16 m depth, 1 m spacing) with alternating expansion agent charging, achieving a 6 m reduction in roof collapse intervals, effectively mitigating overhang hazards. Results confirm that static stress intervention reshapes the roof stress field, inducing tensile failure along predetermined paths without relying on pre-cracks. The findings provide theoretical and technical insights for roof stability control in high-gas coal mines. Full article

This study investigates the failure of a 750A dual-insulated pipeline, where cracks developed along the weld joints during heat supply resumption at the district heating facility. A comprehensive analysis was conducted through visual inspection, mechanical testing, microstructural characterization, finite element analysis (FEA), and electrochemical corrosion testing. The results indicate that cracks were generated in the heat-affected zone (HAZ), primarily caused by galvanic corrosion and thermal expansion-induced stress accumulation. Open circuit potential (OCP) measurements in a 3 M NaCl solution confirmed that the HAZ was anodic, leading to the most vulnerable position to corrosion. Furthermore, localized electrochemical tests were conducted for respective microstructural regions within the HAZ. The results reveal that coarse-grained HAZ exhibited the lowest corrosion potential, giving rise to preferential corrosion, promoting pit formation, and serving as initiation sites for stress concentration and crack propagation. FEA simulations demonstrate that pre-existing microvoids in the HAZ act as stress concentration sites, undergoing a localized stress exceeding 475 MPa. These findings emphasize the importance of controlling microstructural stability and mechanical integrity in welded pipelines, particularly in corrosive environments subjected to thermal stresses. Full article

The instability and collapse of surrounding rock in mine goaf areas often lead to the destabilization of geological structures, surface subsidence, and mining safety accidents. To investigate the evolutionary mechanisms and precursor characteristics of rock instability and failure processes, uniaxial loading and cyclic loading–unloading tests were conducted on red sandstone using a rock mechanics loading system. These experiments aimed to explore the mechanical behavior of the rock and the development process of internal fractures. The characteristics of infrasonic signals generated during red sandstone fracturing and the laws governing damage evolution were analyzed with an infrasonic acquisition system. The research results indicate that the infrasonic signal activity generated by rock under loading conditions can be characterized by three distinct stages, namely the relative stability period, the active period, and the pre-failure precursor period. Prior to peak strength, a substantial number of infrasonic signals are generated in rocks with significant activity; this characteristic is independent of the loading path but dependent on the stress magnitude. The variation in cumulative infrasonic energy reflects the accumulation of damage in rock specimens during the loading process, and as damage accumulates, the stress–strain curve exhibits hysteresis effects and nonlinear increases, accompanied by a rapid rise in infrasonic energy. By analyzing the characteristics of infrasonic parameters and characterizing the damage and its evolutionary features in red sandstone based on infrasonic energy, the internal crack damage evolution process in rocks can be effectively characterized. This approach provides theoretical foundations and technical support for early warning and monitoring prior to rock failure. Full article

Deep grouting rock engineering is faced with the dual influence of high temperature and dynamic load, which has become a hot issue in geotechnical engineering. This study analyzes the mechanical responses and failure properties of deep-grouted fractured rock under real-time coupling of temperature and dynamic loads through the high-temperature-split Hopkinson pressure bar (HT-SHPB), high-speed imaging, and scanning electron microscopy (SEM) tests. Key findings reveal that (1) the dynamic compressive strength of grouted fractured rock exhibits significant temperature dependency, where the strength increases with the increase of temperature, which has been verified by relevant references. From indoor temperature to 100 °C, the dynamic strength increases moderately, while a pronounced increase is observed between 100 °C and 300 °C. (2) In contrast, the dynamic peak strain demonstrates a two-stage evolution, which sharply rises from indoor temperature to 100 °C, followed by a slowly rise from 100 °C to 300 °C. (3) Macroscopically, impact fractures preferentially initiate as parallel lines at the extremities of pre-existing grouted fractures, consistent with stress concentration patterns under dynamic loading. Microscopic analysis reveals that grouting materials effectively suppress micro-crack generation and propagation at 300 °C, attributed to thermally enhanced cementation and pore-filling effects, explaining the variation of the macroscopic dynamic strength with temperature from the microscopic point of view. Full article

In an era of increased need for underground tunnel excavation to address growing urban population and traffic concerns, complaints resulting from blasting vibrations and the frequent execution of uneconomically inefficient blasting operations due to excessive overbreak have become more prevalent. Therefore, it is necessary to develop blasting methods that can reduce blasting vibrations and minimize overbreak. Various patterns of crack induction holes were placed between the presplitting holes to facilitate the formation of controlled pre-cracks to address the limitations of the presplitting blasting method in this study. The author conducted full-scale experimental blasting at a railway tunnel site and analyzed the blasting effects of the crack induction hole method and pre-splitting technique. As a result of the field test, the pre-formed cracks effectively attenuated vibrations generated in the cut blasting area, reducing blasting-induced vibrations by from 9.3% to 33.5%. Additionally, the amount of overbreak was decreased by from 17.9% to 20.2%. Therefore, the use of crack induction holes and pre-splitting blasting methods in underground tunnel blasting is expected to reduce overbreak, thereby lowering reinforcement costs and minimizing vibrations, preventing damage to adjacent structures. This is expected to enable economically and safely executed tunnel blasting operations both directly and indirectly. Full article

Structural degradation in liquid environments can hinder the applications of polymer composites as structural materials. In this work, we study the impacts of methanol on surface cracking and the propagation of pre-formed cracks in UV-irradiated poly(methyl methacrylate)/functionalized graphene (PMMA/FG) composites, followed by the uptake of three different crack-generated solvents, namely 1-butanol, cyclohexanol, and 2EA, respectively. The density of surface cracks increases with the increase in the uptake of the crack-generated solvent. The dependence of the nominal diffusivity for the surface cracking on temperature follows an Arrhenius-like law. The methanol in the composites enhances the uptake of the crack-generated solvent, accompanied by the desorption of methanol, and accelerates the initiation and propagation of surface cracks. The activation energy for the initiation of surface cracks shows an increasing dependence on the Hansen solubility distance from methanol. The progression of the pre-formed crack length with time follows a parabolic law. The nominal diffusivity of the crack-generated solvent for the propagation of the single-crack is greater in the healing zone than in the crack-free zone; the corresponding activation energies exhibit an opposite trend. Increasing the fraction of functionalized graphene and decreasing the UV-irradiation dose cause increases in the energy barriers that need to be overcome for the surface cracking and propagation of preexisting cracks. Full article

Weakly supervised crack segmentation aims to create pixel-level crack masks with minimal human annotation, which often only differentiate between crack and normal no-crack patches. This task is crucial for assessing structural integrity and safety in real-world industrial applications, where manually labeling the location of cracks at the pixel level is both labor-intensive and impractical. Addressing the challenges of labeling uncertainty, this paper presents CrackCLIP, a novel approach that leverages language prompts to augment the semantic context and employs the Contrastive Language–Image Pre-Training (CLIP) model to enhance weakly supervised crack segmentation. Initially, a gradient-based class activation map is used to generate pixel-level coarse pseudo-labels from a trained crack patch classifier. The estimated coarse pseudo-labels are utilized to fine-tune additional linear adapters, which are integrated into the frozen image encoders of CLIP to adapt the CLIP model to the specialized task of crack segmentation. Moreover, specific textual prompts are crafted for crack characteristics, which are input into the frozen text encoder of CLIP to extract features encapsulating the semantic essence of the cracks. The final crack segmentation is determined by comparing the similarity between text prompt features and visual patch token features. Comparative experiments on the Crack500, CFD, and DeepCrack datasets demonstrate that the proposed framework outperforms existing weakly supervised crack segmentation methods, and the pre-trained vision-language model exhibits strong potential for crack feature learning, thereby enhancing the overall performance and generalization capabilities of the proposed framework. Full article

According to the mechanical characteristics of joints in steel–concrete composite bridge decks under the combined bending and shear, improved joint details with simple structure and convenient construction were studied, including lapped U-bars, lapped headed bars, and lapped hook bars. In order to test the mechanical properties of the three joint details and compare them with the existing lapped/welded linear bars, the tests of five specimens were carried out. The cracking load, ultimate load, failure mode, crack pattern, and reinforcement strain were analyzed. The test results showed that the joint with lapped U-bars and hook bars exhibited ductile failure, while the joint with lapped headed and lapped/welded linear bars exhibited brittle failure. The cracking load of the five specimens was basically the same. The crack first occurred at the interface of pre-cast concrete and wet joints. When the ultimate bearing capacity was reached, the vertical main cracks were generated near the interface of the lapped U-bars, lapped hook bars, and welded linear bars specimens. The diagonal cracks were generated at the wet joint of the lapped headed bars specimen and lapped linear bars specimen. The lapped U-bars specimen had the highest bearing capacity, which was 22.8%, 14.2%, 50.4%, and 32.1% higher than the capacities of the lapped headed bars, lapped hook bars, lapped linear bars, and welded linear bars specimens, respectively. The load–deflection curves and crack mode obtained from the FEM of the lapped U-bars joint specimen were consistent with the test results. The bearing capacity of the FEM (351.3 KN) was 1.8% less than the test result (357.6 KN), which indicates that the bearing capacity calculated by the finite element model is reliable. There are 80 models with varying lap lengths and concrete strengths. The self-organizing migrating algorithm was used to fit the coupling effect of lap length and concrete strength. Based on doubly reinforced beam flexural capacity formulas, a bearing capacity calculation for lapped U-bars joint was proposed. The mean value of the formula calculation result and the finite element result ratio is 1.03, and the variance is 0.0004. Full article

The unique properties of graphene have attracted the interest of researchers from various fields, and the discovery of graphene has sparked a revolution in materials science, specifically in the field of two-dimensional materials. However, graphene synthesis’s costly and complex process significantly impairs researchers’ endeavors to explore its properties and structure experimentally. Molecular dynamics simulation is a well-established and useful tool for investigating graphene’s atomic structure and dynamic behavior at the nanoscale without requiring expensive and complex experiments. The accuracy of the molecular dynamics simulation depends on the potential functions. This work assesses the performance of various potential functions available for graphene in mechanical properties prediction. The following two cases are considered: pristine graphene and pre-cracked graphene. The most popular fifteen potentials have been assessed. Our results suggest that diverse potentials are suitable for various applications. REBO and Tersoff potentials are the best for simulating monolayer pristine graphene, and the MEAM and the AIREBO-m potentials are recommended for those with crack defects because of their respective utilization of the electron density and inclusion of the long-range interaction. We recommend the AIREBO-m potential for a general case of classical molecular dynamics study. This work might help to guide the selection of potentials for graphene simulations and the development of further advanced interatomic potentials. Full article

The existence of cracks is a key factor affecting the strength of concrete. However, traditional numerical methods still have some limitations in the simulation of crack growth in fissured concrete structures. Based on this background, the numerical treatment method of particle failure in smoothed particle hydrodynamics (SPH) is proposed, and the generation method for concrete meso-structures under the smoothed particle hydrodynamics (SPH) framework is developed. The concrete meso-models under different pre-existing micro-fissure inclinations and bridge angles (the inner tip line of the double pre-existing micro-fissure is defined as a bridge, and the angle between the bridge and the horizontal direction is defined as the bridge angle) were established, and numerical simulations of the crack propagation processes of concrete structures under tensile stress were carried out. The main findings were as follows: The concrete meso-structures and the pre-existing micro-fissures all have great impacts on the final failure modes of concrete. The stress–strain curve of the concrete model presents four typical stages. Finally, the crack initiation and propagation mechanisms of fissured concrete are discussed, and the application of smoothed particle hydrodynamics (SPH) in crack simulations of fissured concrete is prospected. Full article