Search Results (3,709)

To address the high labor intensity, low level of automation, and insufficient uniformity of spray coverage in plant protection operations for trellised crops in multi-span greenhouses, an autonomous spraying system based on magnetic navigation and RFID node identification was developed. The system mainly includes a differential-drive chassis module, a control module, and a spraying module, with an STM32 microcontroller serving as the core controller. The CAN bus was used for information exchange between modules. Magnetic navigation was employed for continuous path tracking between crop rows, while RFID was used to identify key operation nodes. A fuzzy PID algorithm was adopted to correct path deviations by adjusting the motion states of the differential-drive wheels online. The system thereby achieved autonomous path tracking, steering, obstacle avoidance, and point stopping. To evaluate the performance of the system, tests were conducted on the trellised watermelon in a multi-span greenhouse. The results showed that, at an angular velocity of 0.6 rad/s, the average actual turning angle was 88.48°, corresponding to the smallest steering error. At a travel speed of 0.5 m/s, the average obstacle-avoidance braking distance and RFID point-stopping error were 7.50 cm and 14.44 cm, respectively. The spraying tests showed that, within the tested ranges, spray coverage and coverage uniformity exhibited clearer changes with travel speed than with spray pressure. Among the nine tested operating conditions, the highest observed mean canopy coverage was 40.98% at 0.8 m/s and 0.4 MPa, with a coverage coefficient of variation of 26.56%. These results demonstrate that the developed system can perform autonomous navigation, node-based control, and spraying operations in multi-span greenhouse environments. The findings provide a reference for automated spraying operations of trellised crops in greenhouse environments. Full article

Polygonatum sibiricum Red. is a medicinal and edible homologous plant whose rhizome quality determines its clinical efficacy and commercial value. However, conventional evaluations typically treat the entire rhizome as homogeneous, overlooking the intrinsic heterogeneity along age nodes, and the regulatory mechanisms governing such age-dependent quality variation remain largely unclear. In this study, four-year-old rhizomes were precisely segmented into four age nodes (HJ1–HJ4), and the contents of polysaccharides, flavonoids, and saponins were systematically determined alongside comparative transcriptome sequencing to dissect the age-dependent bioactive variation and underlying molecular mechanisms. The three bioactive components exhibited markedly asynchronous accumulation patterns, with polysaccharides peaking at HJ3, flavonoids at HJ4, and saponins at HJ1. Transcriptomic analysis identified 19,638 differentially expressed genes (DEGs), with KEGG enrichment pointing to starch and sucrose metabolism, flavonoid biosynthesis, and steroid biosynthesis as the pivotal pathways. A coordinated transcriptional reprogramming emerged across development, characterized by early upregulation of backbone-construction genes and progressive activation of modification-and-decoration genes—yet it manifested through pathway-specific regulatory logics, with Mantel tests further confirming that key genes, including β-amylase3, CYP75B11, and squalene epoxidase (SQLE), were significantly correlated with the accumulation of polysaccharides, flavonoids, and saponins, respectively. These findings challenge the conventional assumption that older nodes are universally superior and establish that age-node differentiation reflects fundamental metabolic reprogramming, supporting a node-targeted utilization strategy encompassing HJ3 for polysaccharide-based products, HJ4 for antioxidant applications, HJ1 for saponin extraction, and HJ2 for bulk materials, while providing both mechanistic insights and practical guidance for precision harvesting and quality-oriented management of P. sibiricum. Full article

Automated shiitake pre-grading in forest-understory production environments is challenged by non-uniform illumination, cluttered backgrounds, occlusion, fine-grained appearance differences, and limited edge-computing resources. This study proposes LGB-YOLOv11n, a lightweight detector that assigns modules to feature levels according to their processing roles. Large-kernel separable attention is placed at the P4/16 backbone level and P3/8 detection branch to strengthen contour, texture, edge, and local-deformation representation. Ghost modules compress the P4/16 and P5/32 detection branches, while BiFPN-inspired weighted fusion replaces four neck fusion nodes. On a held-out test set, the model achieved 97.51% mAP@0.5 and 86.62% mAP@0.5:0.95, improvements of 1.21 and 5.62 percentage points over YOLOv11n, respectively. It used 2.39 M parameters, 7.72% fewer than the baseline. Gains were retained in all three leave-one-acquisition-date-out folds, and three random-seed runs yielded 86.49% ± 0.09% mAP@0.5:0.95. Cross-domain single-class evaluation on 222 external images yielded 89.52% AP@0.5 and 63.71% AP@0.5:0.95, compared with 86.73% and 59.84% for YOLOv11n. On an NVIDIA Jetson Orin NX with TensorRT FP16, LGB-YOLOv11n achieved 134.3 ± 6.73 FPS with a mean inference pipeline latency of 7.47 ± 0.39 ms. These results demonstrate a favorable accuracy–complexity trade-off and support real-time model inference for edge-based shiitake appearance pre-grading. Full article

Bullying victimization is consistently associated with depressive symptoms among adolescents, yet the psychological processes underlying this association remain insufficiently understood, particularly among multicultural adolescents. This study tested a theoretically specified serial mediation model involving acculturative stress and ego-resiliency and further explored item-level conditional associations among the study variables. Cross-sectional data were obtained from 1338 multicultural adolescents in South Korea who participated in the fifth wave (2015) of the Multicultural Adolescents Panel Study (MAPS). Serial mediation analysis (PROCESS Model 6) was performed to examine the mediating roles of acculturative stress and ego-resiliency. A Gaussian graphical network was subsequently estimated to identify central nodes, bridge nodes, and node predictability among bullying victimization, acculturative stress, ego-resiliency, and depressive symptoms. Bullying victimization was positively associated with depressive symptoms. Significant indirect associations were observed through acculturative stress and ego-resiliency, both independently and through the theoretically specified serial pathway. In the network, physical aggression (BV4), appearance-related humiliation (BV6), and peer exclusion (BV1) occupied relatively central positions. Rumor spreading (BV5) showed the highest bridge expected influence, whereas BV4 and BV6 showed the highest bridge strength. Acculturative-stress items demonstrated the highest average node predictability. Bullying victimization was associated with depressive symptoms among multicultural adolescents in South Korea, with significant indirect associations involving greater acculturative stress and lower ego-resiliency. Serial mediation and network analyses provided complementary construct- and item-level perspectives, but the cross-sectional findings should be interpreted as associational rather than causal. Full article

We propose a lightweight cone-based Region of Interest (ROI) filtering method for camera–LiDAR fusion on distributed edge systems. Multiple Neural Processing Unit (NPU) nodes perform camera inference, and a central edge board combines their detections with LiDAR point clouds. The relative rotation is obtained from IMU quaternions under a common attitude reference and aligned sensor axes, while camera Field of View (FOV) parameters define the viewing rays. The method avoids a separate offline extrinsic-rotation estimation procedure, but it requires initial alignment, a measured translation vector, and timestamp-based synchronization. Because the rotation follows from the attitude streams rather than from a per-pair calibration session, a camera node can be added or re-aimed without a new calibration session, which lowers the setup cost of extending the system to further viewpoints. A cone membership test replaces four plane-normal dot products with a forward sign test and a squared angular cosine comparison that reuse the same axis–point dot product; on the same hardware, the mean per-camera ROI-filtering and clustering latency decreases from 6.02 to 4.46 ms, a 25.9% reduction. An adaptive threshold tightens the ROI boundary using the angular separation between neighboring detections. Across six overlap events in a parking scenario, pair-level separation succeeds in 2/6 cases (33.3%) with Pyramid and 5/6 cases (83.3%) with Cone+Adp. These preliminary results indicate improved ROI point selection for the tested configurations, rather than a general increase in intrinsic spatial separation capability. Full article

Fault diagnosis in grid-connected grid-forming hybrid energy storage station (GFM-HESS) systems is challenging because fault transients are jointly affected by converter control dynamics, multi-source electrical couplings, operating-condition variations, and measurement noise. To address these characteristics, this paper proposes a virtual synchronous generator (VSG)-constrained physics-informed heterogeneous temporal graph neural network (VSG-PI-HTGNN) for multi-class fault diagnosis. Based on VSG control characteristics and electrical relationships, 18-dimensional node-level features and eight-dimensional global physical features are constructed from ten monitored signals. These signals are further represented as a heterogeneous graph with five node types and eight predefined relation types, while node-type-specific transformations, heterogeneous graph convolution, learnable relation-scaling factors, and a primary–auxiliary dual-output framework are integrated for feature learning. A MATLAB/Simulink electromagnetic transient model is established to generate 3200 samples covering normal operation and nine fault conditions. At a signal-to-noise ratio (SNR) of 18 dB, the proposed model achieves 97.25% test accuracy and a macro-F1 score of 0.9727. Ablation results show that removing all physical information reduces the accuracy to 89.83%, while, under the unified experimental setting, the proposed model obtains higher values of the reported diagnostic metrics than the seven considered benchmark methods. Further evaluations show that the accuracy remains between 95.50% and 98.75% across SNR levels of 10–30 dB and reaches 95.38% with only 20% of the training data. Validation using an independently acquired hardware-in-the-loop (HIL) dataset further yields 92.75% accuracy and a macro-F1 score of 0.9282. Overall, these results indicate that the proposed method provides favorable diagnostic accuracy, noise robustness, data efficiency under limited-sample conditions, and simulation-to-HIL transferability under the evaluated conditions. Full article

Aiming at the problems of high energy consumption, high carbon emissions from data centers and the difficulty of renewable energy accommodation in distribution networks driven by rapid growth in computing tasks, this paper proposes a collaborative trading method for data center–power grid–energy storage systems to improve spatiotemporal flexibility. Firstly, an integrated mechanism model including IT equipment, HVAC cooling systems, delay-tolerant batch tasks and UPS energy storage is established to quantify multi-dimensional internal flexible regulation potential. Secondly, an improved k-means algorithm is adopted for scenario reduction of wind–PV outputs, and a stochastic-robust collaborative trading optimization model considering carbon emission cost is constructed. Multiple practical constraints are incorporated, including power balance, power flow limits, nodal voltage bounds, task service latency and state of charge limits of energy storage. An improved particle swarm optimization with premature-convergence indicator is developed to solve this nonlinear, non-convex, mixed-variable problem. Simulations are carried out on a modified IEEE 33-node test system over a 24 h scheduling horizon. Numerical results demonstrate that compared with the conventional demand-response strategy, the proposed method reduces total operational cost by 10.7%, curtails wind–PV abandoned power, and achieves 28.6% peak-shaving ratio for data center load. Monte Carlo repeated experiments indicate that the improved Particle Swarm Optimization (PSO) reaches a 95% feasible solution rate with an average computation time of 26.8 s for day-ahead dispatch, which satisfies practical engineering requirements. Full article

Satellite-altimetry-derived marine gravity models often exhibit limited regional adaptability and region-dependent residual errors relative to shipborne observations, particularly in coastal, shelf, and slope areas. This study proposes a regional marine gravity refinement method that integrates region-adaptive background-field fusion with multi-relational graph neural network residual learning in the northern South China Sea and adjacent waters. Four background models—SIO/UCSD, SDUST2022GRA, NSOAS24, and SWOT05—were used together with shipborne gravity, bathymetry, distance-to-coast, and survey-line information. A region-adaptive initial field was first constructed according to the error characteristics of the background models in different subregions, and its difference from the shipborne observations was taken as the residual-learning target. The matched shipborne points were then represented as graph nodes, with spatial, terrain, model-response, survey-line, and regional relations used to construct a multi-relational graph. Spatially disjoint blocks were used to separate the training, validation, and test samples. The Multi-relational GNN predicted local residuals, which were added back to the region-adaptive initial field to obtain the refined gravity anomalies. On the held-out test set, the proposed method achieved an RMSE of 6.10 mGal, an MAE of 3.83 mGal, a 95th-percentile absolute error of 13.08 mGal, a bias of −0.19 mGal, and a squared Pearson correlation coefficient r² of 0.945, outperforming the interpolation, machine-learning, and neural-network baselines. Ablation experiments showed that the survey-line relation provided the largest individual contribution. However, the present validation is limited to spatially held-out samples within the existing shipborne survey network; generalization to completely unseen cruises, other marine regions, and areas without nearby shipborne constraints remains to be further verified. Overall, the results indicate that combining regional background-model adaptation with structured residual learning can improve regional marine gravity field refinement in spatially heterogeneous environments. Full article

Liquid-nitrogen biobanks need temporary, spatially distributed temperature records, but conventional batteries and radios become unreliable far below their rated operating temperatures. We developed a finite-duration wireless logging system that separates sensing and communication in both space and time. A remote PT1000 probe follows the cryogenic environment, whereas a polytetrafluoroethylene (PTFE)/aerogel enclosure delays cooling of the battery and electronics; data are stored locally during exposure and retrieved by Bluetooth Low Energy only after warm-up. This differentiated thermal-path and staged-communication architecture is the principal novelty of this work. A transient node model reproduced the internal cooling trend, with a mean absolute error (MAE) of 3.74 °C, a root mean square error (RMSE) of 3.92 °C, and r = 0.9986, with a 123.5 s difference in the time to reach −50 °C. Nine nodes logged for 38.4–47.2 min (mean 44.2 min), and all reconnected after 35 min of warm-up at approximately 25 °C. In a one-node indoor engineering test, historical records were recovered without packet loss over 2–10 m, and 18 locations in an operating biobank yielded retrievable temperature histories. The system therefore supports short, non-real-time mapping and workflow assessment; it is not a substitute for fixed real-time alarm or metrological monitoring systems. Full article

To address redundant and tortuous paths generated for tracked mowers in unstructured orchard environments and the limited adaptability of fixed-gain controllers to nonuniform-curvature paths, this study proposes a joint optimization method combining an improved rapidly exploring random tree (IRRT) planner with fuzzy PD (F-PD) tracking control. The planner uses an obstacle-density-based adaptive step size to balance search efficiency and obstacle-avoidance safety, an improved artificial potential field to bias random samples toward the goal, and cubic B-spline smoothing to generate continuous paths. Based on a differential-steering kinematic model, the F-PD controller uses heading error and its rate of change as inputs and adjusts proportional and derivative gains online through fuzzy inference. Across the three simulated environments, the average reductions in path length, node count, and computation time achieved by IRRT relative to conventional RRT were 10.6%, 11.7%, and 66.7%, respectively. The maximum lateral error of F-PD was 0.43 m, versus 1.42 m for PID. Field tests showed that the IRRT–F-PD combination reduced cumulative operation time and cumulative relative fuel consumption by 30.5% and 26.6%, respectively, compared with RRT–PID. The proposed method improves planning efficiency and curved-path tracking for autonomous orchard mowing. Full article

Predicting progression from mild cognitive impairment (MCI) to Alzheimer’s disease (AD) requires models that represent both regional brain abnormalities and their evolution across repeated examinations. We developed a longitudinal graph neural network that integrates structural magnetic resonance imaging, FDG-PET, regional imaging biomarkers, and clinical covariates across irregular follow-up visits. The study included 614 participants with baseline MCI from the Alzheimer’s Disease Neuroimaging Initiative: 218 converters to AD within five years and 396 non-converters, with 2438 eligible longitudinal visits. Each visit was represented as an 82-node brain graph based on the Desikan–Killiany atlas. Node features combined a 128-dimensional multimodal convolutional embedding with four regional biomarkers. Graph-attention layers modelled spatial dependencies, a node-wise gated recurrent unit modelled longitudinal dependencies, and masked temporal self-attention aggregated variable-length visit sequences. Participants were divided at the subject level into development and held-out test sets, and hyperparameters were selected by five-fold cross-validation within the development set. On the held-out test set of 123 participants, the model reached an area under the receiver operating characteristic curve of 0.859 (95% CI 0.795–0.915), balanced accuracy of 0.805 (95% CI 0.736–0.862), sensitivity of 0.781, and specificity of 0.832. The AUC was numerically higher than that of the strongest baseline, a CNN–GRU sequence model, which reached 0.832 (95% CI 0.758–0.894); the paired AUC difference was 0.027 (95% CI 0.009–0.098), the unadjusted DeLong p-value was 0.026, and the Holm-adjusted p-value was 0.052, which was not significant at the conventional 0.05 threshold after correction for multiple comparisons. In ablation experiments, removing temporal modelling reduced the AUC to 0.818, and removing the spatial graph structure reduced it to 0.808, the two largest reductions observed. Integrated-gradient analysis placed the highest importance on hippocampal and entorhinal regions. Combining graph-based spatial modelling with recurrent longitudinal reasoning was associated with higher discrimination than sequence modelling alone, though this difference was not statistically significant after correction for multiple comparisons. Validation was restricted to a single research cohort (ADNI), and no independent external dataset was used; prospective external validation on an independent cohort is required before the model can be considered for clinical use Because FDG-PET was unavailable for 19.3% of visits, we report the headline result separately from a sensitivity analysis restricted to participants with complete FDG-PET at every visit (development set cross-validated AUC 0.891 vs. 0.874 for the full cohort with masked missing FDG-PET); multimodal performance should be read as cohort-dependent rather than as a single unconditional figure. Full article

Today, with the emergence of the Internet of Things (IoT) in smart home technology, new security issues have arisen, including device diversity, data privacy concerns, and susceptibility to advanced, complex attacks. Traditional Intrusion Detection Systems (IDS) face critical challenges, including single points of failure, high communication overhead, and inherent privacy violations, which the proposed decentralised approach addresses. In this paper, a novel Federated Anomaly Detection for Heterogeneous Smart Home Environment-based Zero-Trust Micro-Segmentation (CloudShield-IoT) framework is introduced. The proposed architecture introduces three key innovations: (i) the dynamic micro-segmentation engine to enforce zero-trust policies by periodically authenticating devices and monitoring their behaviour, (ii) the privacy-preserving federated learning module for collaborative intrusion classification across distributed smart home nodes without exchanging raw data. Note that the core detection module performs supervised intrusion classification using labelled attack categories; the term “anomaly detection” in this paper refers to the broader system-level behavioural-deviation monitoring achieved through the hierarchical trust-scoring mechanism. Moreover, (iii) the hierarchical trust-scoring module adaptively isolates compromised devices in real time. The framework is extensively tested on publicly available benchmark datasets, including N-BaIoT and IoT-23, and compared against 10 state-of-the-art baselines. CloudShield-IoT achieves 98.8% detection accuracy, an F1-score of 0.982, and an AUC of 0.993 without differential privacy, with an inference latency of only 12.4 ms. With practically meaningful (ε = 5.0, δ = 1 × 10⁻⁵) differential privacy via a moments accountant, the framework obtains an accuracy of 97.6%, an F1-score of 0.975, and an AUC of 0.989. The framework demonstrates strong resilience against Byzantine adversarial attacks. It maintains excellent scalability for deployments of up to 500 nodes, supporting heterogeneous device configurations ranging from 10 identical-type devices to 100 mixed-type devices per node. Full article

The quantification of structural resilience and sub-percentile tail risk represents a major challenge across both corporate financial engineering and modern industrial logistics. Traditional aggregation architectures, such as linear risk matrices and parametric copulas, can exhibit computational and sensitivity challenges when modeling extreme tail-risk dependencies under sparse data regimes. While parametric copulas are highly effective under standard conditions, they can be sensitive to parameter specifications and sample size limitations in deep-tail regions. This paper highlights a numerical limitation of the Gumbel extreme-value copula in deep-tail regions (F ≥ 0.999). Analytical results indicate that the logarithmic structure of the tail generator produces progressively higher sensitivity near the distribution boundary, yielding an empirical condition number greater than 1220 at the regulatory 99.9% Value-at-Risk (VaR) threshold. This numerical conditioning issue increases sensitivity to sample noise and data scarcity, resulting in a 36.5% underestimation of systemic tail damage. The proposed model formalizes risk scenarios by mapping multi-node threats as normalized directional unit vectors within a compact 3D vector space. Interactions are then calculated algebraically using geometric invariants—the Dot Product for root-cause convergence and the Cross Product Norm for dynamic, second-order risk resonance—effectively contracting high-dimensional combinations into a stable framework. Rather than treating risks as frame-dependent scalar probabilities, this generalized High-Dimensional Geometric Invariant Operational Risk Framework extends legacy structures with domain-agnostic invariants capturing dynamic risk resonance and multi-trigger cascades. Simulation results across rugged operational environments, acute data scarcity (Ntrain = 100), and high-dimensional scaling (50 risk factors) demonstrate that the proposed model outperforms standard alternatives by a factor of approximately 13 in out-of-sample predictive accuracy (MSE = 0.08193) while maintaining absolute parametric stability. Furthermore, a Taylor-series tensor contraction successfully collapses 1275 second-order interactions into just 2 free parameters. This framework bypasses iterative Maximum Likelihood Estimation (MLE) bottlenecks, unlocking real-time, low-latency Monte Carlo stress testing for systemic banking compliance and global supply chain risk governance. Full article

Environmental-monitoring systems often bundle sensing, communication, storage, visualization, and control into one application, making later changes difficult. We designed a service-oriented platform that separates these functions through defined interfaces. It combines a Raspberry Pi gateway, a dedicated motor-control microcontroller, five environmental sensor modules, Node-RED middleware, a database, and a web interface. Deterministic code alone evaluates threshold and composite rules and controls safety-relevant alerts; an optional large language model (LLM) turns pre-computed statistics and rule outcomes into narrative reports. We examined data acquisition and rule processing during two short indoor campaigns. In the residential campaign, the SCD41 yielded 78 valid three-minute bins (234 min of recorded data) across four sessions between 09:18 and 17:12 local time; binned CO₂ concentrations ranged from 679 to 1471 parts per million (ppm). Using the initial campaign for development and the residential campaign as a temporal holdout, the persistence model produced a 15 min forecast mean absolute error of 58.3 ppm and a root mean square error of 78.8 ppm. A separate controlled experiment generated 270 reports from nine deterministic synthetic scenarios. Every reporter preserved all deterministic alert identifiers, while the fixed template and seven of the nine locally hosted LLMs achieved 100% numerical fidelity. Qwen 3.5 9B was the only LLM that returned all required measured content without automated claim-review flags and produced identical outputs across repetitions for every scenario. These results confirm integration and functional separation under the tested conditions, but they do not demonstrate week-scale reliability, longer-horizon forecasting accuracy, robotic mobility performance, or load scalability. Full article

Modern distribution networks increasingly face reverse power flow, heavy loading or overloading, and voltage violations as distributed generation and flexible demand introduce large spatiotemporal variations in active-power injections and withdrawals. This paper proposes a planning-operation consistent DC-AC time-series optimal power flow (OPF) method for flexible-resource planning and operation optimization to mitigate these problems. The DC module optimizes investment decisions with embedded DG curtailment and flexible-load regulation to improve the operational relevance. The AC module further considers resource reactive-power flexibility and optimizes their operation under voltage constraints. The consistent design of the two modules in objective structure, operating constraints, and flexible-resource representation allows the planning results to be parsed as the initial schedule for AC operation refinement, improving operation-optimization efficiency. Furthermore, the model introduces discrete type-and-number BESS planning, endogenous initial state of charge (SOC) optimization, and a unified flexible-load model to improve operability and economic relevance. The method is implemented in a CloudPSS-based DSLab environment and tested on a real distribution feeder and the IEEE 123-node benchmark. The real-feeder case demonstrates coordinated mitigation of reverse-power export, branch overloads, and voltage violations. In the IEEE 123-node benchmark, the 8760 h AC operation case converges in 376.31 s, confirming tractability for long-horizon time-series optimization. Full article