Search Results (20,950)

Online flatness detection of aircraft rudder surfaces serves as a pivotal core procedure for ensuring the manufacturing precision, aerodynamic performance and operational safety of aeronautical components. Traditional plane fitting-based detection approaches are constrained by low detection efficiency, susceptibility to local optimal solutions, weak anti-noise robustness and limited automation capability, which fail to satisfy the micron-level high-precision online detection requirements for curved composite rudder surfaces in batch manufacturing scenarios. To address the aforementioned technical bottlenecks, this study proposes a bidirectionally coupled PSO-SA hybrid optimization algorithm for non-convex minimum zone flatness evaluation of curved rudder surfaces, which overcomes the unidirectional open-loop iteration limitation inherent in conventional serial PSO-SA composite frameworks. Two targeted algorithmic improvements are elaborated in this work: a residual-adaptive nonlinear inertia weight strategy, which dynamically balances global exploration and local exploitation capabilities based on the fluctuation characteristics of free-form surface measurement residuals; and a measurement noise-modified Metropolis acceptance criterion, which substantially enhances the algorithm’s anti-interference performance against on-machine trigger sampling noise. Integrating with the trigger-type on-machine detection hardware of computer numerical control (CNC) machine tools, an integrated online detection system is established to realize the full-process functions of point cloud data acquisition, error compensation, intelligent plane fitting and flatness error evaluation. Meanwhile, the complete technical workflow involving measurement path planning, probe calibration and algorithm iterative solution is systematically illustrated. Comparative simulation experiments implemented on the MATLAB platform demonstrate that the proposed algorithm exhibits superior performance in convergence speed, fitting accuracy and optimization stability over five mainstream algorithms, including standard particle swarm optimization (PSO), standard simulated annealing (SA), comprehensive learning PSO (CLPSO), adaptive cooling SA and conventional serial PSO-SA. On-machine physical measurement experiments are conducted on 24 aircraft rudder workpieces covering aluminum alloy skins and assembled riveted components. After multi-dimensional systematic calibration, the overall detection error of the developed system is controlled within 1 μm. The experimental results indicate that the average flatness error calculated by the proposed bidirectionally coupled PSO-SA algorithm is 29.7 μm, which is 30.1% and 38.5% lower than that of standard PSO and standard SA, respectively, fully complying with the aviation flatness tolerance specification of 0.1–0.3 mm. Moreover, the full detection cycle for a single workpiece is only 2.1 min, achieving a 34.4% reduction in detection time compared with standard PSO and effectively improving the efficiency of online in-process inspection. One-way analysis of variance (ANOVA) combined with Tukey’s posthoc test further verifies that the accuracy superiority of the proposed algorithm is statistically significant. This research provides a targeted theoretical basis and complete engineering implementation scheme for intelligent flatness detection of aerospace curved thin-walled parts, and offers a valuable technical reference for form and position error evaluation of irregular industrial components under noisy measurement conditions. Full article

This paper provides a comprehensive overview of the main empirical, equivalent-fluid, poroelastic, resonant, and numerical models used to describe sound absorption in porous materials. Each model is described in detail with regard to its theoretical basis, governing equations, and key physical parameters. Special attention is devoted to the assumptions underlying each model, such as whether the frame is rigid or flexible, the applicable frequency range, and the types of porous media they most accurately represent. The advantages and limitations of the different approaches are critically assessed in terms of prediction accuracy, computational complexity, physical interpretability, and experimental requirements. In addition, this paper summarizes and critically discusses published model–experiment comparisons for representative porous and resonant acoustic materials. These comparisons highlight the strengths and weaknesses of different modeling strategies in various acoustic applications and provide guidance for selecting the most suitable model according to the material properties and target frequency range. The review shows that equivalent-fluid models generally provide the best compromise between prediction accuracy and computational efficiency for rigid-frame porous materials, whereas Biot-type poroelastic models are more suitable when frame motion cannot be neglected. Full article

Motor Imagery (MI) decoding based on electroencephalogram (EEG) is promising for Brain–Computer Interface (BCI) applications, yet existing methods generally suffer from two major limitations. (1) Global EEG signal analysis overlooks region-specific neural activity patterns, causing biased feature extraction and poor inter-subject generalization. (2) Few MI-EEG decoding studies adopt frequency decomposition for multi-rhythm feature extraction. Even when adopted, conventional methods rely on predefined frequency bands and suffer from mode mixing, failing to preserve the inherent fractal self-similarity and nonlinear characteristics of EEG signals, restricting the extraction of fine-grained specific features. To address these issues, we propose RMF-Net, a novel model integrating brain region division and Multi-variable Variational Mode Decomposition (MVMD). The model partitions EEG into functional brain regions based on MI neural mechanisms, performs dynamic modal feature extraction for each region via MVMD, and enables efficient cross-regional spatiotemporal feature interaction through an adaptive fusion. On the BCI Competition IV 2a open EEG MI dataset, our model achieves 80.06% accuracy in cross-session tasks and 63.05% in cross-subject tasks, outperforming other mainstream methods. Further analysis verifies that the cross-regional feature weight distribution of RMF-Net conforms to neuroanatomical principles. This work demonstrates that the spatiotemporal feature fusion framework combining brain region segmentation and fractal-aware multimodal signal decomposition is effective for EEG MI decoding tasks. Full article

Underwater object detection is crucial for marine ecological monitoring and resource exploration. However, low underwater contrast causes severe aliasing between foreground and background in the spatial domain, and conventional methods struggle to effectively decouple their features. Transformer architectures incur high computational overhead. Conversely, over-compressing these models severely degrades their ability to detect heavily camouflaged or small marine organisms. High-quality underwater samples are limited, and existing time-consuming generative strategies are hard to implement efficiently on edge devices. To address these challenges, this paper proposes a resource-efficient framework for underwater object detection. First, we design a Wavelet-Enhanced Feature Pyramid Network that combines a saliency-focus mechanism and a discrete wavelet transform to overcome background noise in both spatial and frequency domains, extracting features of hidden small objects. Second, a data-dependent dynamic token pruning technique removes redundant tokens, effectively mitigating the computational bottleneck without sacrificing essential semantic capacity. Finally, for extreme sample scarcity, we introduce a Feature Correction Module and a two-stage fine-tuning and feature correction strategy, using a high-precision teacher model to guide a compressed student network in adaptively compensating for optical shifts with few samples. Experiments on URPC2020 and DUO demonstrate that our method improves small object detection accuracy while reducing parameter count and computational overhead, striking a good balance between accuracy and inference efficiency. Full article

Classical canonical correlation analysis becomes numerically unstable when the number of variables is large relative to the sample size and is sensitive to contamination in observations or individual cells. This study develops an integrated robust and regularized procedure that combines bounded cellwise wrapping, shrinkage estimation of the joint correlation matrix, and robust reweighting in a low-dimensional canonical score space. The resulting observation weights enter a second regularized canonical correlation fit, so the final estimator remains well defined when the combined number of variables exceeds the sample size. The simulation study shows that relative estimation accuracy depends on the signal strength, contamination mechanism, and dimensional configuration. The proposed estimator is competitive in several moderate-signal settings and has a clear computational advantage, whereas the minimum regularized covariance determinant plug-in estimator provides lower estimation error in many high-signal configurations. An additional ultra-high-dimensional experiment demonstrates numerical feasibility with modest memory use but also reveals substantial attenuation, identifying a limitation of the present dense estimator. The results therefore support a regime-dependent interpretation rather than a claim of uniform superiority. The complete reproducible simulation workflow is provided. Full article

Reliable remaining useful life (RUL) estimation is important for the safe and efficient use of lithium-ion (Li-ion) batteries in electric vehicles (EVs) and energy-storage systems. Most data-driven RUL models are trained under controlled laboratory conditions, but their performance can weaken during field operation, where usage, sensing quality, and degradation paths are less predictable. This study proposes a physics-informed laboratory-to-field (L2F) deep learning framework for battery RUL prediction under limited or unavailable target labels. The framework combines three components: a six-channel laboratory cycle representation comprising voltage, current, temperature, cumulative charge throughput, cumulative energy throughput, and voltage derivative; a gated Transformer–Temporal Convolutional Network (TCN) Fusion backbone for modeling long-range and local degradation patterns; and a staged adaptation policy based on paired-view consistency and covariance alignment. The Fusion backbone achieved the lowest held-out XJTU laboratory root mean square error (RMSE) of 46.67, compared with 48.27 for TCN and 48.85 for the Transformer. In the Tsinghua University (Tsinghua) deployment experiment, measured target RUL labels were unavailable after preprocessing and window construction. Therefore, the direct field-side mean absolute error (MAE), RMSE, and coefficient of determination $R^2$ were not computed. The Tsinghua results are interpreted as an unlabeled deployment-credibility and trajectory-regularity assessment, showing operational continuity, finite vehicle-specific predicted trajectories, and reduced local trajectory volatility after staged adaptation. The S2a + S2b policy reduced Fusion RUL volatility from 8.376 to 0.632. In the XJTU laboratory source representation, Integrated Gradients showed that physics-aware channels contributed 30.98% of the attribution mass, increasing from 19.50% in early-life windows to 31.86% in late-life windows. These attributions explain the laboratory six-channel waveform model and are not used as direct evidence of Tsinghua field-feature importance. Full article

The agro-industrial complex of Kazakhstan faces increasing pressure to improve the efficiency of marketing activities under conditions of limited budgets, regional heterogeneity, seasonal demand, and uneven digital infrastructure. Traditional marketing planning methods often rely on expert judgment or single-objective optimization and therefore cannot adequately address the conflicting objectives of maximizing marketing effectiveness, expanding audience coverage, and minimizing campaign expenditure. This study proposes a hybrid decision-support framework that combines the Non-dominated Sorting Genetic Algorithm II (NSGA-II) with machine learning-based clustering for multi-objective marketing budget optimization in the agro-industrial sector. The model considers five major promotion channels: digital media, television, radio, print media, and events. NSGA-II is used to generate Pareto-optimal budget allocation strategies, while K-means clustering classifies the resulting solutions into interpretable strategy groups. The framework was evaluated using synthetic data and enterprise-level data reflecting marketing conditions in Kazakhstan. Results show that strategies combining digital media and television achieve the highest marketing effectiveness and audience coverage, although they require greater campaign expenditure. Radio and print media remain relevant for cost-sensitive strategies, particularly for enterprises with limited resources. Cluster analysis identified three main strategy groups: cost-efficient, balanced, and high-performance configurations. The findings confirm that marketing budget allocation in the agro-industrial sector should be treated as a multi-objective optimization problem rather than a single-criterion decision. The proposed framework provides both computational optimization and managerial interpretability, supporting data-driven, adaptive, and resource-aware marketing planning for agro-industrial enterprises in Kazakhstan. Full article

Electroencephalography (EEG) provides a non-invasive means of supporting Attention-Deficit/Hyperactivity Disorder (ADHD) assessment. Nonetheless, existing pipelines often rely on handcrafted descriptors, segment-wise decisions, or deep architectures with limited subject-level generalization. This work introduces Hidden Markov Model-Induced Stationary RKHS Distance Learning (HIS), a probabilistic framework that represents each subject by a Hidden Markov Model with Gaussian-mixture emissions trained directly from frontal EEG recordings. Rather than vectorizing model parameters, each HMM is mapped to its induced stationary observation distribution and embedded into a Reproducing Kernel Hilbert Space (RKHS), where pairwise subject similarities are computed through a closed-form Hilbert embedding distance. These similarities are subsequently exploited by precomputed-kernel classifiers for subject-level prediction. The proposed method was evaluated against the Probability Product Kernel baseline using both a controlled synthetic EEG benchmark and a public pediatric ADHD dataset under progressively more rigorous validation protocols, culminating in repeated nested cross-validation with bootstrap confidence intervals and permutation testing. On the synthetic benchmark, HIS achieved 95.0% held-out accuracy and consistently outperformed the baseline across classifiers. On a real EEG dataset with 121 subjects, the primary evaluation protocol yielded a balanced accuracy of 73.5% (95% CI: 69.8–77.0%), an AUC of 79.6%, and an MCC of 0.483 (permutation p < 0.001) using an SVM with compact subject-specific HMMs. Complementary hyperparameter analyses and t-SNE visualizations demonstrated that HIS induces more stable and discriminative subject representations than the baseline. These results establish stationary RKHS embeddings of subject-specific HMMs as a leakage-aware framework for EEG-based ADHD decision support and underscore the critical influence of statistically rigorous evaluation protocols on reported classification performance. Full article

The mechanical behavior of calcareous sand differs significantly from that of conventional quartz sands, leading to challenges in offshore geotechnical engineering applications. This distinctive response is closely associated with the complex three-dimensional morphology of calcareous sand particles. However, existing characterization methods are often limited to specific morphological scales and cannot fully describe the multiscale complexity of particle shape. To address this issue, this study performs a comparative morphological analysis of calcareous sand (CS) and Fujian quartz sand (FS) across three particle-size ranges by integrating X-ray micro-computed tomography with spherical harmonic (SH) analysis. Individual particles are reconstructed using SH representation, and a multiscale morphology characterization framework is developed by decomposing particle morphology into three distinct scale levels: large-scale form represented by sphericity, medium-scale angular features represented by roundness, and small-scale surface texture represented by roughness. The results demonstrate that CS and FS exhibit distinct morphological characteristics across different scales, while particle-size effects remain less pronounced within the investigated range. Furthermore, the SH amplitude spectra reveal statistically self-similar characteristics of particle surfaces, allowing the fractal dimension to be correlated with multiscale morphological descriptors. The proposed framework provides a quantitative description of complex particle morphology across multiple scales and may facilitate further investigations of particle-scale mechanical behavior in granular materials. Full article

Audio-visual emotion recognition (AVER) is central to affective computing systems that require reliable, real-time interpretation of human emotions. However, many existing multimodal models treat feature learning and deployment efficiency separately, limiting their ability to preserve hierarchical facial relationships, capture long-range speech dynamics, and operate with low latency in distributed settings. This study proposes a latency-aware hybrid Transformer–capsule network for audio-visual emotion recognition in a simulated edge–fog–cloud environment. The visual stream employs a CNN–Capsule branch to retain spatial hierarchies in facial expressions, while the audio stream uses a CNN–Transformer branch to learn local spectral patterns and long-range temporal dependencies from speech. A cross-modal Transformer fusion module integrates complementary emotional cues, and a latency-aware task-allocation mechanism allocates preprocessing, inference, and training-related operations across edge, fog, and cloud layers according to workload, node capacity, and communication delay. Unlike approaches that optimize multimodal representation learning and distributed deployment as separate problems, the proposed framework adopts a deployment-aware co-design in which spatial visual representation, temporal acoustic modeling, multimodal interaction, and deterministic latency-aware task allocation are coordinated within a unified processing pipeline. The framework is evaluated on RAVDESS, CREMA-D, and SAVEE using a subject-independent protocol. Experimental results show an average accuracy of 91.5%, an F1-score of 90.7%, an MCC of 0.894, and an AUC of 0.950. The framework further incorporates a deterministic latency-aware task-allocation mechanism for coordinating operations across edge, fog, and cloud resources. Physical-device deployment and comprehensive resource profiling remain subjects for future validation. Full article

Non-healing wounds remain a clinical challenge due to their complex pathophysiology and limited therapeutic options. These conditions are driven by complex molecular mechanisms, including inflammation, cell proliferation, and tissue remodeling. Melatonin (MLT) and its N1- and N2-substituted derivatives are known to exhibit potent antioxidant and anti-inflammatory properties; however, their specific therapeutic mechanisms in wound healing remain largely unexplored. Therefore, this study aimed to investigate the potential wound-healing properties of MLT and its six derivatives using an integrated computational and in vitro validation approach. Potential targets of MLT and its derivatives were screened using SwissTargetPrediction (version 2023 release) and SuperPred (version 3.0), yielding 491 candidate targets. These targets were cross-referenced with the GeneCards (version 5.24.0) database to map their involvement across the four phases of wound healing: hemostasis, inflammation, proliferation, and remodeling. Network interaction models were constructed using Cytoscape (version 3.10.3) and GeneMANIA (version 3.6.0), and pathway enrichment was analyzed using the ShinyGO (version 0.85.1) platform. Enrichment analysis prioritized HIF-1-related signaling as a candidate regulatory axis associated with the predicted targets of melatonin derivatives across the inflammatory, proliferative, and remodeling phases of wound healing. In vitro validation using normal human dermal fibroblasts (NHDFs) demonstrated that all compounds, at non-toxic concentrations, significantly enhanced cell viability, as measured by the MTT assay. Furthermore, wound scratch assays revealed that the N2-bromobenzoyl-substituted derivative (EBMLT) accelerated cell migration, achieving complete wound gap closure within 24 h and outperforming the parent compound. Molecular docking simulations using AutoDock 4.2 predicted favorable binding interactions of the derivatives toward key wound healing-related targets (NF-κB, EGFR, VEGFR-1, MMP-1, and MMP-13). Aromatic-substituted derivatives (BMLT, BBMLT, and EBMLT) exhibited more favorable predicted binding interactions than the parent compound across all targets, whereas the aliphatic-substituted derivative (SMLT) showed weaker predicted interactions, particularly with VEGFR-1. These findings suggest that N1- and N2-aromatic substitutions are associated with more favorable binding interactions. Notably, the N2-bromobenzoyl derivative (EBMLT) exhibited the most potent wound-closure activity, highlighting it as a candidate compound for wound-healing applications. Full article

The acoustic frequency fingerprint (AFF) for drone recognition has emerged as a flexible and practical approach to address low-altitude airspace security concerns. However, three problems restrict its real-world application: (1) efficient AFF features are distributed across a wide range of frequencies, making them hard to extract accurately; (2) environmental noise originating from diverse sources, across different bands, cannot be denoised by a general method and limits monitoring accuracy; and (3) multi-band noise greatly increases the labeling workload and computational resources, limiting the flexibility in practical monitoring. This paper proposes a lightweight but efficient acoustic recognition system. We adopt a DE-filter and HLDWT to adaptively extract multi-band AFF features and limit other bands’ noises; a D-MPGAN to decouple and denoise motor and propeller characteristics, avoiding noisy sample labeling; and a lightweight MP-ResNet for motor and propeller feature handling, masking and comprehensive recognition decisions. Furthermore, we build a real dataset. Trained on clean 10 dB data and validated on −30 dB to 10 dB data, our system achieves 85% recognition accuracy at −30 dB with an inference time of 0.058 s/2000, outperforming mainstream methods trained on mixed noisy signals in anti-noise capability and efficiency, thus enabling reliable practical monitoring for low-altitude security. Full article

This work presents a power-friendly framework for efficient manufacturing test of dynamic voltage and frequency scaling (DVFS)-based 3D Systems-on-Chip (SoCs). The proposed approach employs a through-silicon via (TSV)-based inter-layer test architecture to enable compact and high-speed delivery of test data across stacked dies, while a bus-based space- and time-division multiplexing (SDM/TDM) mechanism distributes test data within each layer. Based on this architecture, a test access mechanism (TAM) optimization method is introduced to reduce TSV usage, improve bandwidth utilization, and minimize test application time. The optimization process uses effective pruning criteria to limit the exploration of inefficient TAM configurations while preserving promising design alternatives. In addition, advanced test-scheduling methods are developed to exploit TDM-based parallelism and flexibility, while explicitly enforcing power constraints at the SoC, layer, and voltage-island levels. These scheduling methods combine fast heuristic construction with metaheuristic optimization techniques to improve solution quality without excessive computational cost. Experimental evaluation on artificial DVFS-based 3D SoC instances demonstrates that the proposed framework achieves reductions in TSV count and effectively limits the overall test time compared with baseline approaches. The results confirm that jointly considering TAM design, DVFS-aware scheduling, and power constraints provides an effective and scalable solution for testing complex DVFS-based 3D SoCs. Full article

Objective: To examine survival outcomes and identify clinicopathological factors associated with survival in patients with advanced-stage endometrial cancer who received neoadjuvant chemotherapy before interval debulking surgery (NACT-IDS). Methods: A single-center retrospective cohort study was conducted of patients who were diagnosed with advanced-stage (2009 FIGO IIIB, IIIC, IV) endometrial cancer (2012–2024) and underwent NACT-IDS. Tumor response to NACT was determined with computed tomography and the RECIST criteria, and demographic, clinicopathologic, perioperative, and tumor molecular features (mismatch repair protein [MMR] status and p53 pattern) were collected from medical chart review. Association between tumor molecular features and response to NACT was determined. Primary endpoints were progression-free and overall survival, analyzed with univariate Cox and stratified Kaplan–Meier analysis. Results: Of 42 consecutive patients (median age of 68 years), the majority (n = 26; 61.9%) had a partial tumor response to NACT, with only five (11.9%) having a complete response, four (9.5%) having stable disease, and seven (16.7%) having progressive disease. Most cases had no residual tumor after IDS (n = 35; 83.3%). MMR protein status was associated with the tumor response to NACT (p = 0.013), but p53 status was not. During follow-up, 24 patients died (57.1%), and 31 (73.8%) died or had disease progression. Tumor response to NACT and resection margin status were associated with progression-free survival in unadjusted analyses, but no covariate adjustment was possible with limited sample size. Conclusions: This study highlights MMR-deficiency, response to NACT, and surgical resection status as clinicopathologic features of interest for future studies of prognostic factors and alternative therapies in advanced-stage endometrial cancer. Full article

The soft Bankart lesion is characterized by the abnormal translation of the humeral head during dislocation, which places excessive stress on the labrum and causes it to stretch along with other structures that provide joint stability. This lesion is described as a purely soft tissue injury and occurs due to the detachment of the anteroinferior labroligamentous complex. This research aimed to evaluate the computational biomechanics of a biomodel of the shoulder joint with a soft Bankart lesion during pure abduction using the Finite Element Method (FEM). It evaluates the tissue mechanics of the structures with the lesion, such as ligaments, the articular capsule, and the labrum, which guide and limit the bones of the joint during movement. A healthy shoulder joint biomodel is developed for comparison. The results of stress and strain in the healthy shoulder and the soft Bankart biomodel are analyzed. The results for the soft Bankart biomodel show an increase in stress on the MGHL, with the pIGHL assuming a primary stabilizing role. The CHL and articular capsule limit the excessive displacement of the humeral head. Full article