Search Results (8,111)

Rolling bearings represent one of the core functional components of rotating machinery, with their application scope continuously expanding into various sectors of modern social production and life, making the research on fault diagnosis of rolling bearings increasingly significant. Effective vibration feature extraction and improved classification models are crucial to achieving accurate and automated fault diagnosis of rolling bearings. We proposed a fault diagnosis approach based on a Swin Transformer–Improved ResNet module. In the data preprocessing stage, the frequency-domain features and time-domain multi-scale features of fault signals are extracted using FFT and VMD methods, respectively. And then, dual-channel feature extraction is employed using both the Swin Transformer and Improved ResNet module, followed by feature fusion through an ECA module, thereby enhancing diagnostic accuracy and model robustness. The architecture retains shallow-level feature details while incorporating global contextual information, improving feature representation and detection precision. Extensive experiments were carried out on data collected from an SEU bearing dataset, including model validation, ablation analysis, comparative evaluation and simulated noise testing. An average classification accuracy of 99.41% was achieved by the proposed model under uniform experimental conditions, as evidenced by the obtained experimental results, outperforming other models by at least 0.96%. Even under severe noise interference with a signal-to-noise ratio of -4, the model maintained an average accuracy of 91.92%, exceeding that of noise-resistant counterparts. Moreover, generalization experiments on the CWRU bearing dataset under varying load conditions revealed an average fault diagnosis accuracy exceeding 98%, confirming the model’s strong cross-domain adaptability. Full article

Fusarium head blight (FHB), caused by Fusarium graminearum, is a devastating wheat disease leading to significant yield loss and mycotoxin contamination. This study elucidated the biocontrol mechanism of Bacillus velezensis 20507 fermentation broth against FHB during wheat infection. The broth exhibited strong, time-dependent antifungal activity in vitro, with optimal growth suppression (inhibition rates up to 75%) achieved using broth fermented for 3–7 days. In planta experiments confirmed its efficacy in alleviating disease symptoms. Employing a dual RNA-seq strategy, we analyzed the tripartite interaction between the biocontrol agent, pathogen, and wheat host. Transcriptomic analysis revealed that the broth directly suppressed the pathogen, causing 1510 differentially expressed genes (DEGs, predominantly down-regulated) and disrupting pathways related to carbohydrate metabolism and cell wall integrity. In wheat, the fermentation broth of B. velezensis 20507 counteracted F. graminearum infection by reprogramming the host transcriptome. KEGG analysis during co-inoculation showed that the broth up-regulated defense-related pathways involved in energy, hormone signaling, and cellular maintenance, while down-regulating primary metabolic pathways, indicating a resource reallocation strategy. Furthermore, transcriptomic analysis revealed that the broth alone primed the wheat defense system, and this primed state significantly enhanced the defense response upon pathogen challenge. Untargeted metabolomics identified key antimicrobial compounds, including lipopeptides and the macrolide Macrolactin A. Bioassay-guided fractionation isolated two active fractions (Fr A and Fr B) with potent antifungal activity. This integrated multi-omics study demonstrates that B. velezensis 20507 combats FHB through a coordinated dual mechanism: direct inhibition of the fungus via specific metabolites like Macrolactin A, and simultaneous reprogramming of the host defense and metabolic landscape. These findings provide a scientific foundation for developing this strain as an effective biocontrol agent. Full article

Solid oxide fuel cells (SOFCs) are a promising near-zero-emission propulsion source for Northern Sea Route (NSR) vessels, but their yttria-stabilized zirconia (YSZ) electrolyte and Ni-cermet anode are susceptible to thermomechanical degradation under repetitive start–stop thermal cycling. We develop a hierarchical Bayesian competing-risk framework built on a dual degradation model that decomposes area-specific resistance (ASR) growth into cycle-induced fatigue and time-dependent electrochemical aging and apply it across six NSR duty-cycle scenarios spanning f = 1–27 cycles/month. Posterior inference via the No-U-Turn Sampler (NUTS) yields 17 estimated parameters meeting standard convergence criteria (R̂ ≤ 1.01, ESS_bulk ≥ 479, zero divergent transitions). The analysis identifies a failure-mode transition at f ≈ 3–6 cycles/month: high-frequency routes are crack-dominated (S1a: 10/15 cells fail by crack within the 600-cycle window with 5/15 right-censored), whereas low-frequency routes are ASR-dominated (S3b: 100% ASR). Global sensitivity analysis indicates the time-dependent rate coefficient k_time as the primary remaining-useful-life driver (S_T = 0.37–0.46). Cycle-based maintenance thresholds span 160 cycles (S3b) to ≥600 cycles (S2b), bracketed by S1a (270 cycles, 10.0 months, crack-dominant) and S3a (480 cycles, 160 months, transition regime); qualitative consistency with published experimental data supports physical plausibility. Full article

During flood disasters, on-site operations often face the “triple disruption” of network outages, power cuts and blocked roads. This renders terrestrial cellular infrastructure inoperable and disrupts communication links. Satellite internet can partially restore emergency communications thanks to its wide-area coverage and resistance to ground damage. However, limited computing power, memory and unstable bandwidth at the terminal prevent cloud-based flood segmentation from providing near-real-time situational awareness. This paper therefore proposes a lightweight semantic flood segmentation framework for emergency terminals that uses satellite internet. This comprises a parallel dual-branch design with a lightweight U-Net-style convolutional neural network (CNN) branch for local boundary details and a compact Transformer branch for global context. A dynamic gated fusion mechanism (DGFM) balances local texture and global information adaptively. Experiments on the public synthetic aperture radar (SAR) dataset Sen1Floods11 demonstrate that the hybrid architecture strikes a balance between accuracy and inference efficiency. The proposed method combines gated fusion with quality-aware training. Compared to a lightweight CNN baseline and state-of-the-art segmentation models using the same protocol, the proposed configuration (Hybrid-Gated with Quality-Aware Training) achieves the highest mean intersection over union and F1 score among the compared fusion variants, while maintaining competitive false alarm and risk-sensitive performance under deployment constraints. This aligns with the preferences of emergency decision makers. The framework provides a deployable perception module for emergency systems supported by low-orbit satellites and terrestrial networks under triple-disruption conditions. Full article

This study addresses the problem of robust video-based tracking of laboratory rats in open-field and Y-maze experiments under challenging acquisition conditions, including non-uniform illumination, low contrast, and heterogeneous recording setups. Existing approaches based on classical image processing or deep learning often fail to maintain stable localization under such conditions or require large, annotated datasets. We propose a hybrid tracking framework that combines an improved motion–appearance voting mechanism with consistency-constrained optimization for open-field experiments, together with a comparative deep learning-based detection strategy for Y-maze analysis. The proposed method introduces (i) adaptive dual-threshold motion extraction, (ii) directionally constrained temporal validation, and (iii) a robustness-driven fusion of motion and appearance cues. Experimental results demonstrate that the proposed approach achieves reliable tracking with a maximum localization error below 10 pixels under severe illumination variations. In the Y-maze scenario, a comparative evaluation of multiple detectors (YOLOv5, YOLOv9, YOLO12, Faster R-CNN) highlights the trade-off between accuracy and inference time, with YOLOv9 providing the best balance. The main contribution consists of enabling robust behavioral quantification in low-quality experimental conditions using limited training data, bridging the gap between classical tracking robustness and deep learning flexibility. Full article

This study analyzes whole-body vibration (WBV) exposure of an agricultural tractor operator during three different primary tillage systems: Standard Tillage (ST), Conservation Tillage Deep (CTD), and Conservation Tillage Shallow (CTS). Measurements were conducted in accordance with ISO 2631-1 and ISO 2631-4 along three orthogonal axes (x, y and z) at the operator’s seat. Descriptive and inferential statistical analyses indicate that while none of the mean vibration values exceeded the regulatory limit value of 1.15 m/s² defined in Directive 2002/44/EC, several measurements—particularly in the y-axis during ST (0.715 m/s²)—surpassed the exposure action value of 0.5 m/s². These findings suggest that prolonged daily exposure under similar operational conditions may pose long-term health risks for tractor operators. The highest mean WBV values were recorded in the x- and y-axes during CTS (0.354 m/s² and 0.446 m/s², respectively), whereas the z-axis exhibited the highest values during ST (0.426 m/s²). Conservation Tillage Deep (CTD) demonstrated the most favorable vibration profile in the vertical axis (0.344 m/s²), indicating its potential dual benefit for soil structure preservation and operator ergonomics. Although all measured values remained below the regulatory limit, the frequent exceedance of the action value underscores the importance of exposure time management, regular maintenance of suspension systems, and implement selection as practical mitigation strategies. This comparative assessment provides baseline WBV data for reduced-tillage systems on hydromorphic soils and offers axis-specific guidance for optimizing operator comfort in sustainable mechanization practices. Full article

Action chunking strategies in robot imitation learning struggle to dynamically balance between long-range motion efficiency and short-range operational precision due to their fixed planning horizon. This paper presents an Adaptive Action Chunking framework that enables robots to dynamically predict the optimal action chunk length based on real-time visual context. We design an end-to-end dual-branch network comprising a shared visual encoder, a parallel action prediction head, and a chunk-size prediction head. Experiments on two real-world bimanual robot manipulation tasks (transport-and-place and flip-and-handover) demonstrate that the method autonomously derives two distinct intelligent strategy patterns—phase-aware switching and sustained high-frequency adjustment—in response to task uncertainty. It significantly outperforms fixed-chunk baselines in both success rate and efficiency. Ablation studies confirm that the performance gain stems from the adaptive decision-making mechanism itself. Full article

This paper proposes a finite-time target tracking control for multi-rotor unmanned aerial vehicles (UAVs) based on a vision-inertial-stabilized platform. To address the challenge of stable and accurate moving target tracking, the sliding mode control (SMC) technique is used to overcome limitations of conventional control algorithms, such as poor robustness and slow convergence speed. First, by computing the pixel deviation between the target and the image center, a kinematic model of the tracking target is established. Then, by introducing homogeneous system theory into the sliding mode surface design, a non-singular fast integral terminal sliding mode control (NFITSMC) is designed for target tracking via regulating the rotational angular acceleration of dual actuators in the vision inertial stabilized platform, thereby driving the pixel deviation to converge to zero in a finite time. Strict theoretical analysis is given to prove the finite-time stability and robustness of the closed-loop control system. Furthermore, simulation results demonstrate that the proposed method maintains higher tracking accuracy than SMC, ISMC, and TSMC. Full article

Precise temporal segmentation of acoustic cough signals is critical for digital health, yet existing literature predominantly focuses on simple event detection rather than exact boundary delineation. To bridge this gap, we introduce a comprehensive benchmarking framework specifically designed to systematically evaluate continuous boundary detection performance using modern deep learning architectures. Built upon this evaluation paradigm, we propose a novel Dual-Branch Bi-Mamba architecture that effectively integrates the local morphological feature extraction capabilities of a 2D U-Net with the long-range sequential modeling power of 1D Bidirectional State-Space Models (SSMs). Evaluated on the clinical DKPNet41 dataset, the proposed compact 0.54-million-parameter model achieved an F1-Score of 87.66% while reducing offset boundary error by over 50%. Operating 56× faster than real time on a standard CPU, this study establishes a reliable evaluation framework for precise boundary segmentation and provides a computationally efficient architectural solution for high-resolution automated acoustic signal processing. Full article

Recent advancements in flexible robotic mechanisms have drawn great attention in dual pneumatic artificial muscle (PAM) applications. However, the complex inherent characteristics of dual PAMs, particularly highly nonlinear and time-varying properties, may cause state variables to exceed allowable constraints. Furthermore, the dual PAM control system faces additional challenges from potential singularity issues arising from multiple control variables. To solve the singularity issues and satisfy the overshoot constraints, a novel robust control approach for dual PAMs is suggested. Comparative experimental results on the dual PAM platform are posed to confirm the robustness and efficacy of the suggested control. To our knowledge, this is the first control methodology for PAMs that effectively addresses overshoot limitations and input singularities. Full article

The purpose of this in vitro study was to evaluate the mechanical performance, dimensional accuracy, and bonding behavior of fused filament fabrication (FFF)-printed provisional restorations made from polymethyl methacrylate (PMMA) and polyethylene terephthalate (PET), and compare them with digital light processing (DLP)-printed and computer-aided numerical control (CNC)-milled ones. Occlusal veneers (OV), posterior crowns (PC), and anterior crowns (AC) (n = 30) were fabricated using FFF (PMMA, PET), DLP (acrylate), and CNC (PMMA) to assess initial fracture load (IFL). To determine reproducibility three restorations of each group were scanned and compared with each other; to determine printing accuracy the scanned restorations were compared with the STL generated for manufacturing. For shear bond strength (SBS) testing, 72 PMMA (FFF) specimens were conditioned with either Monobond Plus (MP) or Visiolink (VL) and bonded with acrylic cylinders using a dual-cure luting composite (Variolink Esthetic DC). Half of each group underwent thermocycling (10,000 cycles, 5 °C/55 °C, 30 s dwell time); the remainder was tested initially. Additionally, 48 FFF-printed PC were fabricated from PET and PMMA to investigate the fracture load in relation to the adhesive material (FL). PMMA crowns were conditioned with MP (n = 16) or VL (n = 16) and bonded with Variolink Esthetic DC. PET crowns were cemented with either Meron (ME) or Ketac Cem Plus (KE). Half of the PMMA and all PET crowns were subjected to masticatory simulation (1,200,000 cycles, 5 N, 5 °C/55 °C, 60 s dwell). Data were analyzed using Kolmogorov–Smirnov, Kruskal–Wallis, and Mann–Whitney U tests, including IFL, SBS and FL parametric tests, and comparisons were carried out using an independent t-test ($\alpha$ = 0.05). FFF-fabricated restorations showed the lowest fracture load values and CNC-fabricated the highest (p < 0.001). OV fabricated via DLP and CNC exhibited the highest fracture load (p < 0.001). For FFF, PC demonstrated the highest values (p < 0.028), whereas AC showed the lowest fracture load values (p < 0.001). VL showed higher initial SBS than MP (p < 0.001) and no impact on aging (p < 0.608). All MP samples showed debonding after thermocycling. Within PET and PMMA, no impact of luting/cement material on fracture load was observed (p = 0.116–0.282). The fracture load decreased after masticatory simulation (MP-PMMA: p < 0.001, VL-PMMA: p = 0.27). DLP-fabricated restorations showed the highest reproducibility and printing accuracy. CNC and FFF-PET showed comparable values. FFF-PMMA showed the greatest deviations. CNC-fabricated provisional restorations exhibited the highest fracture load. AC presented the lowest fracture load values. DLP provided the highest reproducibility and accuracy. VL achieved superior bonding to PMMA surfaces. Thermomechanical aging significantly reduced fracture load in both PET and PMMA restorations, regardless of luting material. Full article

Ilex rotunda Thunb. is a prestigious ornamental tree renowned for its vibrant red fruits, yet the molecular mechanisms governing its fruit color variation remain poorly understood. The discovery of a rare yellow-fruited natural bud sport cultivar, ‘Peace Time’, provides an ideal model to investigate these processes compared to the wild-type red fruit. In this study, we integrated physiological evaluations, untargeted metabolomics, and de novo transcriptomics across multiple fruit developmental stages to elucidate the basis of this color transition. Our results demonstrated that the yellow phenotype is characterized by high lightness and yellowness values, driven by the profound suppression of anthocyanin biosynthesis. Biochemical and transcriptomic profiling revealed that DFR (dihydroflavonol 4-reductase), a critical “gatekeeper” gene, experiences severe transcriptional silencing in the yellow-fruited cultivar. This enzymatic bottleneck triggers a “passive substrate overflow,” redirecting shared precursors toward the parallel flavonol branch, resulting in the substantial accumulation of specific flavonols, including rutin and isoquercitrin. Furthermore, correlation network analysis highlighted a putative dual regulatory module associated with this metabolic reprogramming: the down-regulation of the putative activator bHLH30 coupled with the robust up-regulation of the putative repressor bHLH51, together likely contributing to the silencing of DFR transcription. These findings provide a comprehensive “dual-module” and “passive overflow” framework for fruit coloration in I. rotunda, highlighting a remarkable metabolic plasticity that reshapes this cultivar’s phytochemical profile and offers vital insights for future ornamental breeding. Full article

Phototherapy is a physical treatment modality that utilizes natural or artificial light sources and harnesses radiant energy to treat diseases. Dynamic monitoring of the actual light dose received by tissues is crucial to the success of phototherapy. However, most current phototherapy devices feature bulky and complex hardware and depend on fixed parameters or surface measurements for dose estimation, failing to provide precise, real-time monitoring of light dose distribution that is tailored to individual users, specific treatment sessions, and different body regions. Furthermore, most of these devices are incapable of generating tunable and stable LED light. This study presents a preliminary diffusion equation-based proof-of-concept for a wearable, integrated dual-mode sensing system for real-time dynamic monitoring of tissue light dose and temperature change. The system, controlled by a single-chip microcontroller, rapidly extracts key tissue optical parameters via a custom multi-wavelength LED optical probe and provides real-time feedback on light dose distribution through a dynamic tissue optical simulation model. To expand the monitoring dimensions, the system innovatively integrates a thermal sensor. This sensor enables synchronous monitoring of the temperature field in the treatment area, thereby allowing for an estimation of the combined photothermal effect. The system features a compact design, user-friendly operation, fast and stable communication, and repeatable and reliable detection. With promising clinical application prospects, it holds the potential to evolve into a portable, home-use, safe, effective, wearable, and cost-effective phototherapy device. Full article

Wireless Sensor Networks (WSNs) enable energy-efficient data collection in dynamic environments but continue to face the dual challenges of severely constrained node energy and the spatiotemporal heterogeneity of data traffic. Inspired by honeybee foraging behavior, this paper proposes a hybrid optimization framework that integrates mixed-integer linear programming (MILP) and Markov decision processes (MDP), utilizing Q-learning for adaptive decision-making. The proposed framework systematically maps the dual-layer decision-making mechanism of honeybee foraging onto a synergistic architecture combining MILP-based global planning and MDP-based local adaptation, offering a novel bio-inspired solution for mobile sink trajectory planning and adaptive routing. Specifically, the upper-level MILP module simulates a colony-level global assessment of distant nectar sources, generating an initial global trajectory by determining the optimal access sequence of cluster heads to minimize the movement cost of the mobile sink. The lower-level Q-learning module simulates the individual-level local adaptation, where bees adjust harvesting behavior in real-time based on nectar quality and distance. This module continuously optimizes routing parameters based on real-time network states, including residual energy, the ratio of surviving nodes, data queue lengths, and cluster head density. The algorithm employs an ϵ-greedy strategy to balance exploration and exploitation, while a periodic decision-update mechanism is introduced to harmonize computational efficiency with learning stability. Furthermore, a multi-objective reward function is designed to jointly optimize energy efficiency, network lifetime, end-to-end latency, and path length. Extensive simulation results demonstrate that the proposed MILP-MDP hybrid framework significantly outperforms several representative baseline algorithms in terms of network lifetime extension and energy balance. These findings validate that the integration of bio-inspired foraging strategies and reinforcement learning provides an efficient and robust solution for trajectory planning and adaptive routing in dynamic WSNs. Full article

The demand for functional dairy products is increasing, in response to the adverse correlation between high saturated fat consumption and cardiovascular health problems. The present study investigated the physicochemical and sensory features of a prototype of functional reduced-fat Mozzarella cheese fortified with inulin made from milk obtained by integrating the cattle diet with laurel essential oil (LEO). Two samples were compared over a 10-day storage period: a whole-milk Mozzarella cheese (MC), and a reduced-fat Mozzarella cheese fortified with 10% (w/v) of inulin (MI). The results show that incorporating inulin during the stretching phase required more time (2.55 min longer) to obtain the final product. However, in addition to a 5% fat decrease, the MI cheese achieved an inulin content of 3.31%, satisfying the European Regulation No 1924/2006, for the “Source of Fiber” claim. On the other hand, from a nutritional perspective, the dietary LEO integration significantly modulated the lipid fraction of the sample, resulting in a 40% increase in monounsaturated fatty acids (MUFAs) and a marked enrichment in polyunsaturated fatty acids (PUFAs). Considering the texture attributes, the incorporation of inulin during the stretching phase led to the formation of a micro-gel that acted as a functional filler, resulting in significantly higher hardness (33.41 N for MI and 16.10 N for MC), throughout the 10-day storage period. Temporal Check-All-That-Apply (TCATA) analysis confirmed that while the MI sample introduced vegetable and cooked milk notes, MI maintained major textural integrity throughout the shelf-life. These findings demonstrate that the synergy between inulin fortification and dietary laurel essential oil supplementation represents a highly effective strategy for producing reduced-fat pasta filata cheeses. This dual approach not only preserves sensory and textural integrity but also yields a high-value functional product characterized by an optimized fatty acid profile and a significant fiber intake. Full article