Search Results (4,071)

This manuscript presents a study of the Stress-Driven integral Model (SDM) for the bending response of Bernoulli–Euler nanobeams. Unlike conventional approaches that reformulate the nonlocal integral problem into an equivalent differential form, a direct numerical strategy is developed to solve the integral equation. The proposed framework enables a systematic comparison of six different convolution kernels (Helmholtz, Gaussian, Lorentzian, triangular, Bessel and hyperbolic cosine), highlighting how their mathematical properties influence the structural response. To address issues related to long-range interactions and the potential ill-posedness of the integral operator, an adaptive regularization procedure based on the Morozov discrepancy principle is introduced. Furthermore, a local clipping and renormalization technique is proposed to properly account for boundary effects while preserving the weighted averaging property of the kernels. Validation against available analytical solutions for the Helmholtz kernel demonstrates high accuracy, with errors below 1%. The results show that the nonlocal parameter significantly affects structural rigidity depending on the kernel shape and that the proposed approach ensures consistent convergence to the local solution as the nonlocal parameter tends to zero. Full article

High-resolution remote sensing image scene classification (HRRSI-SC) is crucial for obtaining accurate Earth surface information. However, the task remains challenging due to significant background interference, high intra-class variation, and subtle inter-class similarities. Convolutional neural networks (CNNs) are constrained by their local receptive fields, which limits their ability to capture long-range spatial dependencies. On the other hand, Vision Transformers (e.g., ViT-B-16) excel at global feature extraction but often suffer from high computational complexity and may lack the inherent inductive biases for local feature modeling that CNNs possess. To address these limitations, this paper proposes a cross-level feature complementary classification framework based on Lie Group manifold space, termed CBCAM-LGM. Within the proposed CBCAM-LGM framework, multi-granularity features are first distilled via a global average pooling layer to suppress redundant information. The core of our approach, the cross-level bidirectional complementary attention module (CBCAM), then enables the adaptive fusion of features from both branches through a cross-query attention mechanism. Furthermore, by employing parallel dilated convolutions and a parameter-sharing strategy, the model captures multi-scale contextual information by sharing a single set of convolutional weights, which reduces the computational complexity to merely 1.21 GMACs while preserving multi-scale representation with minimal parameter overhead. Extensive experiments on challenging benchmarks demonstrate the model’s efficacy, as it achieves a state-of-the-art classification accuracy of 97.81% on the AID, surpassing the ViT-B-16 baseline by 1.63%, while containing only 11.237 million parameters (an 87% reduction). These results collectively affirm that our model presents an efficient solution characterized by high accuracy and low complexity. Full article

Indoor 3D point cloud semantic segmentation is a fundamental task for fine-grained scene understanding and intelligent perception. Due to the prohibitive cost of dense point-wise annotations, weakly supervised learning has emerged as a promising alternative for indoor point cloud segmentation. However, existing weakly supervised methods commonly rely on fixed confidence thresholds for pseudo-label selection, which exhibit limited generalization caused by threshold sensitivity, underutilization of informative low-confidence regions, and progressive noise accumulation during self-training. To address these issues, we propose TGR-T, a weakly supervised framework for indoor 3D point cloud semantic segmentation that incorporates truncated-Gaussian-weighted reliability with adaptive dynamic thresholding. Specifically, a reliability-adaptive dynamic thresholding strategy is introduced to guide pseudo-label selection based on the evolving confidence statistics of unlabeled mini-batches, with exponential moving average smoothing employed to produce stable global estimates and robust separation of reliable and ambiguous regions. To further exploit uncertain regions, a learnable truncated Gaussian weighting function is designed to explicitly model prediction uncertainty within the ambiguous set, providing soft supervision by assigning adaptive weights to low-confidence predictions during optimization. Extensive experimental results demonstrate that the proposed framework significantly enhances the exploitation of unlabeled data under extremely limited supervision: extensive experiments conducted on standard indoor 3D scene benchmarks demonstrate that TGR-T achieves competitive or superior segmentation performance under extremely sparse supervision and can even outperform several fully supervised baselines trained with dense annotations while using only 1% labeled points, thereby substantially narrowing the performance gap between weakly supervised and fully supervised 3D semantic segmentation methods. Full article

Airport agglomeration coupling coordination is a key indicator of healthy regional aviation development. This study constructs an evaluation index system from three dimensions—airport production, infrastructure construction, and network support—and assesses the coupling coordination capability of China’s four major airport agglomerations using the entropy weight method and a coupling coordination model. Furthermore, an Airport Consistency Index is innovatively introduced as the reciprocal of the coefficient of variation, and an overall coordination degree is developed under the framework of “balanced average level + consistency correction.” By incorporating the inverse coefficient of variation, the proposed index explicitly assesses airport agglomeration dispersion in coordination performance, thereby mitigating the risk that a strong performance at leading airports masks structural imbalances within the system. This refinement enhances the diagnostic precision of the overall coordination assessment by integrating both average development level and internal convergence. Based on calculations for 2020–2024, the overall coordination ranking is Beijing–Tianjin–Hebei, Guangdong–Hong Kong–Macao Greater Bay Area, Yangtze River Delta, and Chengdu–Chongqing. The Beijing–Tianjin–Hebei agglomeration shows strong and stable coordination with limited sensitivity to external conditions, whereas the Yangtze River Delta is more environmentally sensitive due to its large number of airports. The Greater Bay Area demonstrates solid coordination with substantial synergy potential, while Chengdu–Chongqing exhibits relatively weak coordination and considerable room for improvement. The proposed model effectively evaluates the overall coordination degree of airport agglomerations and supports targeted development recommendations. Full article

Accurate estimation of vehicle sideslip angle is vital for the stability and safety of four-wheel independent drive electric vehicles (4WIDEVs), but it faces challenges, including model uncertainties caused by tire yaw stiffness variations and system delays. This paper proposes a novel adaptive fusion strategy that combines the dynamic robust observer (DRO) and the improved adaptive square-root unscented Kalman filter (ASUKF). The DRO is designed based on a two-degrees-of-freedom vehicle model and ensures stability through linear matrix inequalities (LMIs), effectively handling parameter uncertainties and time delays; the ASUKF utilizes a three-degrees-of-freedom model and the magic formula tire model, combined with Sage–Husa adaptive filtering, to address the nonlinear tire dynamics. The key innovation of this paper is the introduction of a fuzzy-rule-based adaptive weighting mechanism that dynamically adjusts the fusion weights of the DRO and ASUKF in real time, thereby exploiting their complementary advantages under uncertainty and nonlinear conditions. The simulation and experimental validations demonstrate that this method significantly improves estimation accuracy, reducing the estimation error of vehicle sideslip angle by an average of 9.36%, and maintains robust performance and dynamic adaptability in various conditions, providing a reliable solution for the real-time state estimation of intelligent electric vehicles. Full article

Urban innovation capacity is increasingly critical to city development, and quantitative assessments of innovative cities’ innovation capability can be achieved via the composite index method, which fully integrates multidimensional indicators. This study develops a hierarchical data envelopment analysis (H-DEA) method to establish a composite index evaluation model for innovation capacity, which features flexible and objective two-level indicators—an advantage that avoids subjective weight assignment and adapts well to the hierarchical structure of innovation evaluation indicators. The proposed H-DEA model is applied to evaluate 67 innovative cities in China, yielding composite scores and rankings that are further compared with those from the traditional weighting method. Sensitivity analysis is conducted by adjusting different upper and lower bounds of the H-DEA model to verify its robustness. Additionally, these 67 cities are divided into four regions, with region-specific weights assigned to the evaluation indicators in the model. The results show that the eastern region has the highest average innovation capacity (0.3783), where technological innovation (weight 0.27) serves as a key driving force; the western region has the lowest average innovation capacity (0.3235), and its innovative cities should prioritize improving outcome transformation capacity (weight 0.1357). Overall, technological innovation receives the highest average weight (0.2422), while outcome transformation capacity gets the lowest (0.1647). Full article

Under the background of the “dual carbon” strategy and regional coordinated development, the synergistic evolution of low-carbon logistics, regional economy, and ecological environment in the Yellow River Basin has become a key pathway to achieving high-quality development. Taking nine provinces (autonomous regions) within the basin as the study area, this paper constructed a coupling coordination evaluation index system for the LREES (Low-carbon Logistics–Regional Economy–Ecological Environment System), and measured the comprehensive development level of each subsystem using the entropy weight method. Based on the coupling coordination degree model, the temporal evolution of the three systems from 2010 to 2024 was systematically evaluated. In addition, global and local spatial autocorrelation models were introduced to identify spatial clustering patterns, while the obstacle degree model was used to identify key constraints at both the criterion and indicator levels. The results revealed that: the overall development level of the LREES systems steadily increased, with reduced regional disparities; the coupling coordination degree showed a trend of “fluctuating rise–gradual coordination,” with the average value increasing from 0.450 to 0.623, indicating continuously enhanced synergy; spatially, a gradient pattern of “downstream > midstream > upstream” emerged, accompanied by significant positive spatial autocorrelation; resource endowment and development scale were major constraints, while construction level, operational efficiency, and governance capacity were secondary. High-frequency obstacle indicators included per capita water resources, total import and export volume, and urban sewage treatment capacity. These findings offer theoretical support and policy guidance for promoting green transformation, enhancing system synergy, and advancing coordinated regional development in the Yellow River Basin. Full article

The damage status of shield machine disc cutters directly impacts the safety and efficiency of tunnelling projects. Current manual inspection methods involve high risks and low efficiency, while existing detection methods suffer from low accuracy and poor real-time performance in complex environments, often lacking quantitative analysis capabilities. To address these issues, this paper proposes an intelligent identification and quantitative assessment method for disc cutter damage based on the Deep Forest (DF) model. First, an eddy current sensor calibration platform was established, and a mapping relationship between output voltage and actual wear was developed through piecewise fitting to achieve precise wear quantification. In the data preprocessing stage, signal quality was improved via filtering, and typical damage features such as edge chipping, cracks, and eccentric wear were extracted using pulse edge detection. These feature segments were then resampled to construct the model training dataset. The DF model utilizes a hierarchical ensemble structure to mine data correlations, enabling accurate identification of four states: normal, edge chipping, eccentric wear, and cracks. Simultaneously, a DF regression model was employed to provide continuous quantitative predictions of damage size. Experimental results show that the classification model achieved accuracies of 98%, 96%, and 96% on the training, validation, and test sets, respectively, with weighted average F1-scores exceeding 0.96. The regression model achieved a coefficient of determination (R²) of 0.9940 and a root mean square error (RMSE) of 0.4051 on the test set. Both models demonstrate excellent performance and generalization, achieving full coverage from “qualitative state identification” to “quantitative wear assessment,” thereby providing reliable decision support for cutter maintenance and replacement. Full article

Examining the spatial differentiation and constraining factors of rural resilience at the micro-scale is essential for navigating compounded risks and unbalanced urban–rural development. The study takes 170 villages in Qianshan City, Anhui Province, as the study sample and constructs a four-dimensional resilience evaluation system encompassing economic, social, infrastructural, and ecological dimensions. The research systematically assesses rural resilience levels and obstacle factors using the entropy weight method, spatial autocorrelation analysis, and the obstacle degree model. The results indicate that: (1) The overall rural comprehensive resilience in Qianshan City is at a moderately low level, with an average value of 0.133, ranging from 0.0604 to 0.4805. Significant inter-village disparities exist, forming a distinct “central agglomeration–peripheral dispersion” spatial pattern driven by urban proximity. (2) The resilience of each subsystem also exhibits pronounced heterogeneity: economic resilience is generally low; infrastructural resilience shows the greatest variation; social resilience is relatively stable in its spatial distribution; and ecological resilience demonstrates a “high in the northwest–low in the southeast” pattern. (3) Hotspots of comprehensive resilience, as well as economic, social, and infrastructural resilience, are concentrated around the central–southern urban areas with stronger development foundations, whereas hotspots of ecological resilience are independently distributed within ecologically advantageous zones. (4) Rural resilience is primarily constrained by deficits in public service accessibility and infrastructure conditions. Notably, the primary obstacle factors exhibit high consistency across villages with different geomorphic conditions. Finally, this study proposes coordinated enhancement strategies for economic development, infrastructure improvement, ecological conservation, and social governance in Qianshan City, providing a scientific basis for rural resilience building and governance. Full article

This article presents an approach to improving additive manufacturing (AM) processes by examining mental workload and human error during component fabrication using fused deposition modeling (FDM). A cross-sectional study involving experts and novice participants was conducted using a multi-stage approach based on Hierarchical Task Analysis (HTA), the NASA TLX Workload, and the Systematic Human Error Prediction Approach (SHERPA). A sample of two experts and two novice participants was studied. The HTA technique analyzed three main tasks: canister disk replacement, software configuration, and nozzle change. Each was divided into several subtasks. The mental workload results indicate that, among expert participants, the NASA-TLX dimensions with the highest average weighted scores were performance (77.5), time demand (65), and frustration (65). Among novice participants, the dimensions with the greatest impact were effort and frustration (90) and mental demand (87.5). The SHERPA method identified 19 human errors: 13 were action errors (68.4%), 3 were verification (checking) errors (15.7%), 2 were recovery errors (10.5%), and 1 was a selection error (5.2%). These results indicate differences in mental workload dimensions between experts and novices that may affect performance and human–machine interaction during additive manufacturing processes. Accordingly, preventive and corrective actions were recommended to minimize errors that can lead to material waste and financial losses. The study also identified low-to-high demand across key dimensions and a substantial number of errors in interactions with AM interfaces. Full article

As a privacy-preserving distributed machine learning paradigm, federated learning (FL) faces serious communication bottlenecks in practical deployment. In this paper, we propose FedLayerPrune, a communication-efficient federated learning method that integrates three synergistic components: (i) a layer-adaptive pruning strategy that dynamically allocates pruning rates based on layer sensitivity and network depth; (ii) a heterogeneity-aware aggregation mechanism that combines sample-size weighted averaging with mask consensus voting to enhance robustness under non-IID data distributions; and (iii) a dynamic pruning rate scheduler that progressively increases compression intensity across training rounds. Unlike existing approaches that apply uniform pruning or consider these techniques in isolation, FedLayerPrune achieves a principled coordination among layer-wise importance evaluation, temporal pruning scheduling, and heterogeneous model aggregation. Extensive experiments on CIFAR-10, MNIST, and Fashion-MNIST demonstrate that FedLayerPrune reduces communication costs by up to 68.3% compared with standard FedAvg, while maintaining model accuracy within a 2% margin. Moreover, our method exhibits stronger robustness and faster convergence under severe non-IID data distributions. These results suggest that FedLayerPrune provides a practical and effective solution for deploying federated learning in resource-constrained edge computing environments. Full article

Gas turbines play a critical role in modern power systems, yet their transient operations (e.g., start-up, load mutation) induce significant thermal inertia in metal components, leading to deviations between simulation results and actual performance. Traditional low-dimensional (1D/0D) simulation models sacrifice detailed flow and temperature field information to reduce computational load, while high-dimensional (3D) computational fluid dynamics (CFD) models are impractical for full-system simulations due to excessive computational costs. This discrepancy creates a critical trade-off between simulation accuracy and efficiency in gas turbine thermal inertia studies. To address this challenge, this study proposes a temperature-gradient-guided dynamic genetic optimization sampling algorithm (TDGA) and integrates it into a multi-dimensional data scaling framework for gas turbines. A fully coupled simulation framework was established, combining 3D CFD models for turbine flow paths (resolving detailed flow and temperature fields) and 1D thermal models for metal components (casing, hub, blades). The TDGA was designed to enable efficient data interoperability between models: it incorporates a dynamic encoding mechanism, temperature gradient weight matrix, density penalty term, quantity penalty term, and regularization term to optimize sampling point distribution. Dynamic weight coefficients for each objective function term and adaptive crossover/mutation probabilities were introduced to balance global exploration (early iterations) and local exploitation (late iterations) during optimization. Comparative analysis showed that the TDGA achieved a mean squared error (MSE) of 15.52K, far lower than those of traditional Latin Hypercube Sampling (75.07K) and Bootstrap Sampling (64.38K). It allocated 70.11% of sampling points to high-temperature gradient regions while reducing the total number of sampling points to 2765. During the middle stage of the gas turbine start-up process, compared with the traditional Latin Hypercube Sampling and Bootstrap Sampling, the average error of the proposed sampling algorithm is reduced by 17.4% and 13.3%, respectively. The proposed TDGA-based framework effectively balances simulation accuracy and computational efficiency, providing a reliable approach for the transient thermal analysis of gas turbines. Full article

Background/Objectives: To evaluate the challenges and limitations of a deep learning model for automated choroidal boundary detection in eyes with geographic atrophy (GA) using spectral-domain OCT (SD-OCT), and to assess the workflow efficiency of an AI-assisted manual verification approach. Methods: In this retrospective study, total 5723 scans (Heidelberg Spectralis) with GA were analyzed. A previously validated tool (NMI ChoroidAI) was used to segment the choroidal inner (CIB) and outer (COB) boundaries. We compared the “AI-assisted” workflow (automated segmentation followed by manual verification) against “manual segmentation only” in terms of accuracy and time consumption. Slice-wise boundary errors were graded as 0 (accurate), 1 (≤33% deviation), 2 (33–66% deviation), or 3 (>66% deviation). Outcomes included error rates and weighted F₁ score (and precision where applicable). Total time for manual-only segmentation versus AI-assisted verification was recorded. -Interreader variability was assessed between the two readers using intraclass correlation coefficient. Results: For CIB, only 5.2% of B-scans showed any deviation (strictly accurate in 94.8%), with weighted F₁ score 0.97 and precision 1.00. COB was more error-prone: 19.0% of B-scans showed deviation; however, when minor deviations were considered acceptable, COB acceptability increased to 94.2% (i.e., 5.8% remained >33% deviated). Only 13.2% of B-scans required minor manual correction. For a 97-scan volume, processing time decreased from an average of 7 h (manual only) to 45 min (AI + human verification), an approximate 90% reduction in manual effort. Inter-reader agreement was high (ICC 0.923 for CIB and 0.938 for COB). Conclusions: Although the deep learning model exhibits limitations in COB detection due to artifacts, it serves as a valuable assistive tool. Our model substantially reduces human effort, but mandatory human verification is required to correct boundary errors caused by hyper-transmission before use in clinical trials. Full article

Accurate altitude estimation is critical for unmanned aerial vehicles (UAVs), yet barometric measurements are susceptible to atmospheric drift and dynamic disturbances. To address these limitations, this paper proposes a dual-layer, real-time differential barometric altimetry framework that integrates a ground reference station with an onboard fusion scheme based on Adaptive Weighted Averaging (AWA) and an Adaptive Extended Information Filter (AEIF). The ground reference station suppresses low-frequency atmospheric variations, while the onboard AEIF incorporates a physical pressure–height model and adaptive noise estimation to maintain a fast dynamic response. The proposed method is validated through numerical simulations, hardware-in-the-loop (HIL) experiments, and real flight tests. In a two-hour outdoor flight test, compared with barometric systems operating without a reference station, the proposed approach reduces the altitude RMSE from 4.05 m to 0.31 m, achieving an approximately order-of-magnitude improvement in representative scenarios and demonstrating decimeter-level altitude measurement accuracy. Full article

The hyperspectral infrared observations of the Geostationary Interferometric Infrared Sounder (GIIRS) on the Fengyun-4A (FY-4A) satellite are an important data source for numerical weather prediction (NWP) assimilation. However, there are systematic differences between observed and simulated brightness temperatures (i.e., the observation increments contain predictable systematic bias components). To address the issue that traditional linear methods struggle to capture the nonlinear relationships between biases and forecast predictors, this study proposes an intelligent bias correction method that integrates ensemble learning and explainable artificial intelligence. First, the entropy reduction method is used to select 69 mid-wave channels. Then, Random Forest, XGBoost, LightGBM, Decision Tree, and Extra Tree are used as base learners to construct a weighted average ensemble model. Training and validation are conducted using high-frequency clear-sky observation data from FY-4A/GIIRS during Typhoon Lekima. The results show that: (1) the ensemble learning correction method outperforms single models and traditional offline methods, with root mean square errors of brightness temperature bias of less than 0.9209 K for the training set and 1.4447 K for the test set; (2) Shapley Additive Explanations (SHAP)-based interpretability analysis reveals the contribution and nonlinear influence mechanisms of factors such as longitude, atmospheric thickness, surface temperature, and total precipitable water on bias correction. This study provides an intelligent bias correction framework with both high precision and explainability, offering a reference for the bias correction and assimilation applications of hyperspectral satellite observations like GIIRS. Full article