Search Results (42,129)

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

The force ripple of a permanent magnet synchronous linear motor (PMSLM) caused by multi-source disturbances in practical applications seriously restricts its high-precision motion control performance. The traditional single-mechanism model has difficulty fully characterizing the nonlinear disturbance factors, while the data-driven method has real-time limitations. Therefore, this paper proposes a hybrid modeling framework that integrates the physical mechanism and measured data and realizes the dynamic compensation of the force ripple by constructing a collaborative suppression algorithm. At the mechanistic level, based on electromagnetic field theory and the virtual displacement principle, an analytical model of the core disturbance terms such as the cogging effect and the end effect is established. At the data level, the acceleration sensor is used to collect the dynamic response signal in real time, and the data-driven ripple residual model is constructed by combining frequency domain analysis and parameter fitting. In order to verify the effectiveness of the algorithm, a hardware and software experimental platform including a multi-core processor, high-precision current loop controller, real-time data acquisition module, and motion control unit is built to realize the online calculation and closed-loop injection of the hybrid compensation current. Experiments show that the hybrid framework effectively compensates the unmodeled disturbance through the data model while maintaining the physical interpretability of the mechanistic model, which provides a new idea for motor performance optimization under complex working conditions. Full article

In recent years, research on plant disease detection has combined artificial intelligence, hyperspectral imaging, unmanned aerial vehicle remote sensing, and other technologies, promoting the transformation of pest and disease control in smart agriculture towards digitalization and artificial intelligence. This review systematically elaborates on the research status of non-destructive detection techniques used for plant disease identification and detection, mainly introducing the following two types of methods: spectral technology and imaging technology. It also elaborates, in detail, on the principles and application examples of each technology and summarizes the advantages and disadvantages of these technologies. This review clearly indicates that non-destructive detection techniques can achieve plant disease and pest detection quickly, accurately, and without damage. In the future, integrating multiple non-destructive detection technologies, developing portable detection devices, and combining more efficient data processing methods will become the core development directions of this field. Full article

Electromagnetic and magnetic devices are increasingly prevalent in sectors such as transportation, industry, and renewable energy due to the ongoing electrification trend. These devices exhibit nonlinear behavior, particularly under signals rich in harmonics. They require precise and appropriate modeling for accurate sizing. Identifying model-specific parameters, which depend on frequency, is crucial. This article focuses on a specific frequency range where a circuit model with series resistance and inductance, along with a parallel resistance to account for iron losses (${\mathrm{R}}_{\mathrm{iron}}$), is applicable. While the determination of series elements is well documented, the determination of ${\mathrm{R}}_{\mathrm{iron}}$ remains complex and debated, with traditional methods neglecting operating conditions such as magnetic saturation. To address these limitations, an innovative experimental method is proposed, comprising two main steps: determining the complex impedance of the magnetic device and extracting ${\mathrm{R}}_{\mathrm{iron}}$ from the model. This method aims to provide a more precise and representative estimation of ${\mathrm{R}}_{\mathrm{iron}}$, improving the reliability and accuracy of electromagnetic and magnetic device simulations and designs. The obtained values of the iron loss equivalent resistance are different by at least 300% than those obtained by an impedance analyzer. The proposed method is expected to advance the understanding and modeling of losses in electromagnetic and magnetic devices, offering more robust tools for engineers and researchers in optimizing device performance and efficiency. Full article

Background: Blue light (BL) irradiation has been shown to induce photobiomodulation (PBM) in cells. Here, we investigate its influence on cell types involved in wound healing. Methods: Cellular responses of immortalized human keratinocytes (HaCaTs), normal human dermal fibroblasts (NHDFs), and human umbilical vein endothelial cells (HUVECs) after light treatment at 450 nm were analyzed by kinetic assays on cell viability, proliferation, ATP quantification, migration assay, and apoptosis assay. Gene expression was evaluated by transcriptome analysis. Results: A biphasic effect was observed on HaCaTs, NHDFs, and HUVECs. Low-fluence (4.5 J/cm²) irradiation stimulated cell viability, proliferation, and migration. mRNA sequencing indicated involvement of transforming growth factor beta (TGF-β), ErbB, and vascular endothelial growth factor (VEGF) pathways. High-fluence (18 J/cm²) irradiation inhibited these cellular activities by downregulating DNA replication, the cell cycle, and mismatch repair pathways. Conclusions: HaCaTs, NHDFs, and HUVECs exhibited a dose-dependent pattern after BL irradiation. These findings broaden the view of PBM following BL irradiation of these three cell types, thereby promoting their potential application in wound healing and angiogenesis. Our data on low-fluence BL at 450 nm indicates clinical potential for a novel modality in wound therapy. Full article

Internal fins are commonly utilized as a passive technique to enhance natural convection, but their efficiency depends on complex interplay between fin design, material properties, and convective strength. This study presents an extensive numerical analysis of buoyancy-driven flow in square cavities containing a single horizontal fin on the hot wall. Over 9000 simulations were conducted, methodically varying the Rayleigh number (Ra = 10 to 10⁵), Prandtl number (Pr = 0.1 to 10), and fin characteristics, such as length, vertical position, thickness, and the thermal conductivity ratio (up to 1000), to assess their overall impact on thermal efficiency. Thermal enhancements compared to scenarios without fins are quantified using local and average Nusselt numbers, as well as a Nusselt number ratio (NNR). The results reveal that, contrary to conventional beliefs, long fins positioned centrally can actually decrease heat transfer by up to 11.8% at high Ra and Pr due to the disruption of thermal plumes and diminished circulation. Conversely, shorter fins located near the cavity’s top and bottom wall edges can enhance the Nusselt numbers for the hot wall by up to 8.4%, thereby positively affecting the development of thermal boundary layers. A U-shaped Nusselt number distribution related to fin placement appears at Ra ≥ 10³, where edge-aligned fins consistently outperform those positioned mid-height. The benefits of high-conductivity fins become increasingly nonlinear at larger Ra, with advantages limited to designs that minimally disrupt core convective patterns. These findings challenge established notions regarding passive thermal enhancement and provide a predictive thermogeometric framework for designing enclosures. The results can be directly applied to passive cooling systems in electronics, battery packs, solar thermal collectors, and energy-efficient buildings, where optimizing heat transfer is vital without employing active control methods. Full article

This systematic literature review explores the socio-ethical implications of Artificial Intelligence (AI) in contemporary media content generation. Drawing from 44 peer-reviewed sources, policy documents, and industry reports, the study synthesizes findings across three core domains: bias detection, storytelling transformation, and ethical governance frameworks. Through thematic coding and structured analysis, the review identifies recurring tensions between automation and authenticity, efficiency and editorial integrity, and innovation and institutional oversight. It introduces the Human–AI Co-Creation Continuum as a conceptual model for understanding hybrid narrative production and proposes practical recommendations for ethical AI adoption in journalism. The review concludes with a future research agenda emphasizing empirical studies, cross-cultural governance models, and audience perceptions of AI-generated content. This aligns with prior studies on algorithmic journalism. Full article

The Mayanghe National Nature Reserve, a key habitat for the endangered François’ langur (Trachypithecus francoisi), faces significant anthropogenic disturbances, including extensive distribution of croplands, roads, and settlements. These human-modified features are predominantly concentrated at elevations between 500 and 800 m and on slopes of 10–20°, which notably overlap with the core elevation range utilized by François’ langur. Spatial analysis revealed that langurs primarily occupy areas within the 500–800 m elevation band, which comprises only 33% of the reserve but hosts a high density of human infrastructure—including approximately 4468 residential buildings and the majority of cropland and road networks. Despite slopes >60° representing just 18.52% of the area, langur habitat utilization peaked in these steep regions (exceeding 85.71%), indicating a strong preference for rugged karst terrain, likely due to reduced human interference. Habitat type analysis showed a clear preference for evergreen broadleaf forests (covering 37.19% of utilized areas), followed by shrublands. Landscape pattern metrics revealed high habitat fragmentation, with 457 discrete habitat patches and broadleaf forests displaying the highest edge density and total edge length. Connectivity analyses indicated that distribution areas exhibit a more continuous and aggregated habitat configuration than control areas. These results underscore François’ langur’s reliance on steep, forested karst habitats and highlight the urgent need to mitigate human-induced fragmentation in key elevation and slope zones to ensure the species’ long-term survival. Full article

This study synthesizes and comparatively investigates two squaric acid-based phthalocyanine-like dyes, SNF and its long-chain alkylated derivative LNF, to systematically elucidate the influence of peripheral hydrophobic groups on their third-order nonlinear optical (NLO) properties. The NLO characteristics were comprehensively characterized using femtosecond Z-scan and I-scan techniques at both 800 nm and 900 nm. Both dyes exhibited strong saturable absorption (SA), confirming their potential as saturable absorbers. Critically, the comparative analysis revealed that SNF exhibits a significantly greater nonlinear absorption coefficient (β) compared to LNF under identical conditions. For instance, at 800 nm, the β of SNF was approximately 3–5 times larger than that of LNF. This result conclusively demonstrates that the introduction of long hydrophobic alkyl chains attenuates the NLO response. Furthermore, I-scan measurements revealed excellent SA performance, with high modulation depths (e.g., LNF: 43.0% at 900 nm) and low saturation intensities. This work not only clarifies the structure–property relationship in these D-A-D dyes but also presents a clear strategy for modulating the NLO properties of organic chromophores for applications in near-infrared pulsed lasers. Full article

Background: Comorbid personality disorders (PDs) in patients with anorexia nervosa (AN) are associated with increased psychopathology, higher suicide risk, and poorer treatment response and outcomes. This study aimed to examine associations between gray matter (GM) volume and PDs in female adolescents with AN before and after short-term psychotherapeutic and nutritional therapy. Methods: Eighteen female adolescents with acute AN, mean age 15.9 years, underwent 3T magnetic resonance imaging before and after weight restoration. The average interval between scans was 2.6 months. Structural brain changes were analyzed using voxel-based morphometry. PDs were assessed using the Structured Clinical Interview for DSM-IV Axis II Disorders (SCID II) and the Assessment of Identity Development Questionnaire. Results: SCID-II total scores showed significant positive associations with GM volume in the mid-cingulate cortex at both time points and in the left superior parietal–occipital lobule at baseline. The histrionic subscale correlated with GM volume in the thalamus bilaterally and the left superior parietal–occipital lobule in both assessments, as well as with the mid-cingulate cortex at follow-up. Borderline and antisocial subscales were associated with GM volume in the thalamus bilaterally at baseline and in the right mid-cingulate cortex at follow-up. Conclusions: PDs in female adolescent patients with AN may be specifically related to GM alterations in the thalamus, cingulate, and parieto-occipital regions, which are present during acute illness and persist after weight restoration therapy. Full article

Addressing the technical demands for the rapid, precise detection of earthquake-triggered landslides in loess tablelands, this study proposes and validates an innovative methodology integrating enhanced deep learning architectures with large-tile processing strategies, featuring two core advances: (1) a critical enhancement of YOLOv8’s shallow layers via a higher-resolution P2 detection head to boost small-target capture capabilities, and (2) the development of a large-tile segmentation–tile mosaicking workflow to overcome the technical bottlenecks in large-scale high-resolution image processing, ensuring both timeliness and accuracy in loess landslide detection. This study utilized 20 km² of high-precision UAV imagery acquired after the 2023 Gansu Jishishan Ms 6.2 earthquake as foundational data, applying our methodology to achieve the rapid detection and precise segmentation of landslides in the study area. Validation was conducted through a comparative analysis of high-accuracy 3D models and field investigations. (1) The model achieved simultaneous convergence of all four loss functions within a 500-epoch progressive training strategy, with mAP50(M) = 0.747 and mAP50-95(M) = 0.46, thus validating the superior detection and segmentation capabilities for the Jishishan earthquake-triggered loess landslides. (2) The enhanced algorithm detected 417 landslides with 94.1% recognition accuracy. Landslide areas ranged from 7 × 10⁻⁴ km² to 0.217 km² (aggregate area: 1.3 km²), indicating small-scale landslide dominance. (3) Morphological characterization and the spatial distribution analysis revealed near-vertical scarps, diverse morphological configurations, and high spatial density clustering in loess tableland landslides. Full article

This paper presents a novel digital identity ecosystem built upon a hierarchical structure of Blockchain tokens, where both government-certified and uncertified tokens can coexist to represent various attributes of an individual’s identity. At the core of this system is the government, which functions as a trusted authority capable of creating entities and issuing a unique, non-replicable digital identity token for each one. Entities are the exclusive owners of their identity tokens and can attach additional tokens—such as those issued by the government, educational institutions, or financial entities—to form a verifiable, token-based digital identity tree. This model accommodates a flexible identity framework that enables decentralized yet accountable identity construction. Our contributions include the design of a digital identity system (supported by smart contracts) that enforces uniqueness through state-issued identity tokens while supporting user-driven identity formation. The model differentiates between user types and certifies tokens according to their source, enabling a scalable and extensible structure. We also analyze the economic, technical, and social feasibility of deploying this system, including a breakdown of transaction costs for key stakeholders such as governments, end-users, and institutions like universities. Considering the benefits of blockchain, implementing a digital identity ecosystem in this technology is economically viable for all involved stakeholders. Full article

Numerical simulations enable us to couple multiphase flow and geochemical processes to evaluate how sequestration impacts brine chemistry and reservoir properties. This study investigates these impacts during CO₂ storage at the San Juan Basin CarbonSAFE (SJB) site. The hydrodynamic model was calibrated through history-matching, utilizing data from saltwater disposal wells to improve predictive accuracy. Core-scale simulations incorporating mineral interactions and equilibrium reactions validated the model against laboratory flow-through experiments. The calibrated geochemical model was subsequently upscaled into a field-scale 3D model of the SJB site to predict how mineral precipitation and dissolution affect reservoir properties. The results indicate that the majority of the injected CO₂ is trapped structurally, followed by residual trapping and dissolution trapping; mineral trapping was found to be negligible in this study. Although quartz and calcite precipitation occurred, the dissolution of feldspars, phyllosilicates, and clay minerals counteracted these effects, resulting in a minimal reduction in porosity—less than 0.1%. The concentration of the various ions in the brine is directly influenced by dissolution/precipitation trends. This study provides valuable insights into CO₂ sequestration’s effects on reservoir fluid dynamics, mineralogy, and rock properties in the San Juan Basin. It highlights the importance of reservoir simulation in assessing long-term CO₂ storage effectiveness, particularly focusing on geochemical interactions. Full article

This study conducts an in-depth analysis of the dynamic characteristics of a single reheat steam turbine generator unit and its speed control system under variable operating conditions. A comprehensive simulation model was constructed to comprehensively evaluate the impact of the heat extraction system on the dynamic behavior of the unit, which integrates the speed control system, actuator, single reheat steam turbine body, and once-through boiler dynamic coupling. This model focuses on revealing the mechanism of the heat extraction regulation process on the core operating parameters of the unit and the system frequency regulation capability. Based on the actual parameters of a 300 MW heat unit in a power plant in Guangxi, the dynamic response of the established model under typical dynamic conditions such as extraction flow regulation, primary frequency regulation response, and load step disturbance was simulated and experimentally verified. The results show that the model can accurately characterize the dynamic characteristics of the heat unit under variable operating conditions, and the simulation results are in good agreement with the actual engineering, with errors within an acceptable range, effectively verifying the dynamic performance of the heat system module and the rationality of its control parameter design. This study provides a reliable theoretical basis and model support for the accurate simulation of the dynamic behavior of heat units in the power system and the design of optimization control strategies for system frequency regulation. Full article

Multimodal sentiment analysis (MSA) faces key challenges such as incomplete modality inputs, long-range temporal dependencies, and suboptimal fusion strategies. To address these, we propose MGMR-Net, a Mamba-guided multimodal reconstruction and fusion network that integrates modality-aware reconstruction with text-centric fusion within an efficient state-space modeling framework. MGMR-Net consists of two core components: the Mamba-collaborative fusion module, which utilizes a two-stage selective state-space mechanism for fine-grained cross-modal alignment and hierarchical temporal integration, and the Mamba-enhanced reconstruction module, which employs continuous-time recurrence and dynamic gating to accurately recover corrupted or missing modality features. The entire network is jointly optimized via a unified multi-task loss, enabling simultaneous learning of discriminative features for sentiment prediction and reconstructive features for modality recovery. Extensive experiments on CMU-MOSI, CMU-MOSEI, and CH-SIMS datasets demonstrate that MGMR-Net consistently outperforms several baseline methods under both complete and missing modality settings, achieving superior accuracy, robustness, and generalization. Full article