Search Results (935)

Accurate photovoltaic (PV) power output forecasting is critical for ensuring stable operation of modern power systems, yet it is constrained by high-dimensional redundancy in input weather data and the inherent heterogeneity of output scenarios. To address these challenges, this paper proposes a novel cascaded data-driven forecasting approach that enhances forecasting accuracy through systematically improving and optimizing the feature extraction, scenario clustering, and temporal modeling. Firstly, guided by weather data–PV power output correlations, the Deep Autoencoder (DAE) is enhanced by integrating Pearson Correlation Coefficient loss, reconstruction loss, and Kullback–Leibler divergence sparsity penalty into a multi-objective loss function to extract key weather factors. Secondly, the Fuzzy C-Means (FCM) algorithm is comprehensively refined through Mahalanobis distance-based sample similarity measurement, max–min dissimilarity principle for initial center selection, and Partition Entropy Index-driven optimal cluster determination to effectively cluster complex PV power output scenarios. Thirdly, a Long Short-Term Memory–Temporal Pattern Attention (LSTM–TPA) model is constructed. It utilizes the gating mechanism and TPA to capture time-dependent relationships between key weather factors and PV power output within each scenario, thereby heightening the sensitivity to key weather dynamics. Validation using actual data from distributed PV power plants demonstrates that: (1) The enhanced DAE eliminates redundant data while strengthening feature representation, thereby enabling extraction of key weather factors. (2) The enhanced FCM achieves marked improvements in both the Silhouette Coefficient and Calinski–Harabasz Index, consequently generating distinct typical output scenarios. (3) The constructed LSTM–TPA model adaptively adjusts the forecasting weights and obtains superior capability in capturing fine-grained temporal features. The proposed approach significantly outperforms conventional approaches (CNN–LSTM, ARIMA–LSTM), exhibiting the highest forecasting accuracy (97.986%), optimal evaluation metrics (such as Mean Absolute Error, etc.), and exceptional generalization capability. This novel cascaded data-driven model has achieved a comprehensive improvement in the accuracy and robustness of PV power output forecasting through step-by-step collaborative optimization. Full article

The acquisition of nighttime remote-sensing visible-light images is often accompanied by low-illumination effects and haze interference, resulting in significant image quality degradation and greatly affecting subsequent applications. Existing low-light enhancement and dehazing algorithms can handle each problem individually, but their simple cascade cannot effectively address unknown real-world degradations. Therefore, we design a joint processing framework, WFDiff, which fully exploits the advantages of Fourier–wavelet dual-domain features and innovatively integrates the inverse diffusion process through differentiable operators to construct a multi-scale degradation collaborative correction system. Specifically, in the reverse diffusion process, a dual-domain feature interaction module is designed, and the joint probability distribution of the generated image and real data is constrained through differentiable operators: on the one hand, a global frequency-domain prior is established by jointly constraining Fourier amplitude and phase, effectively maintaining the radiometric consistency of the image; on the other hand, wavelets are used to capture high-frequency details and edge structures in the spatial domain to improve the prediction process. On this basis, a cross-overlapping-block adaptive smoothing estimation algorithm is proposed, which achieves dynamic fusion of multi-scale features through a differentiable weighting strategy, effectively solving the problem of restoring images of different sizes and avoiding local inconsistencies. In view of the current lack of remote-sensing data for low-light haze scenarios, we constructed the Hazy-Dark dataset. Physical experiments and ablation experiments show that the proposed method outperforms existing single-task or simple cascade methods in terms of image fidelity, detail recovery capability, and visual naturalness, providing a new paradigm for remote-sensing image processing under coupled degradations. Full article

The red soil region in southern China is an ecologically fragile area. Although ecological engineering construction has achieved phased results, there are still obvious gaps in research on the mechanisms underlying vegetation dynamics in response to natural and anthropogenic variables. Changting County (CTC) serves as a typical case of vegetation degradation and restoration in the region. We examined the vegetation dynamics in CTC with the fraction vegetation cover (FVC) based on kernel normalized difference vegetation index-based dimidiate pixel model (kNDVI-DPM) and employed the optimal parameter-based geographical detector (OPGD), multiscale geographically weighted regression (MGWR), and partial least square structural equation modeling (PLS-SEM) to analyze interaction mechanisms between vegetation dynamics and underlying factors. The FVC showed a fluctuating upward trend at a rate of 0.0065 yr⁻¹ (p < 0.001) from 2000 to 2020. The spatial distribution pattern was high in the west and low in the east. Soil and terrain factors were the primary factors dominating the spatial heterogeneity of FVC, soil organic matter and elevation showing the most significant influence, with annual mean q-values of 0.4 and 0.3, respectively. Climate, terrain, and soil properties positively and anthropogenic activities negatively impacted vegetation. From 2000 to 2020, the path coefficient of anthropogenic activities to FVC decreases from −0.152 to −0.045, the adverse effects of human activities are diminishing with ongoing ecological construction efforts. Climate and anthropogenic activities act indirectly on vegetation through negative effects on soils and terrain. The impact of climate on soils and terrain is gradually lessening, whilst the influence of anthropogenic activities continues to grow. This study provides an analytical framework for understanding the complex interrelationships between vegetation changes and the underlying factors. Full article

To address the logistical challenges of traffic congestion and environmental concerns associated with carbon emissions in last-mile delivery, this paper explores the potential of vehicle–drone cooperative delivery. The existing studies are predominantly confined to single-drone scenarios, failing to simultaneously consider the constraints of drone payload capacity and endurance. This limitation leads to task allocation imbalance in large-scale customer deliveries and low distribution efficiency. Firstly, a mathematical model for vehicle–multi-drone collaborative delivery with payload and endurance constraint (VMDCD-PEC) is proposed. Secondly, an improved genetic algorithm (IGA) is developed, as follows: 1. designing a hybrid selection strategy to achieve symmetrical equilibrium between exploration and exploitation by adjusting the weights of dynamic fitness–distance balance, greedy selection, and random selection; and 2. introducing the local search operator composed of gene sequence reversal, single-gene slide-down, and random half-swap to improve the neighborhood quality solution mining efficiency. Finally, the experimental results show that compared with a traditional genetic algorithm (GA) and adaptive large neighborhood search (ALNS), the IGA requires less time to find solutions in various test cases and reduces the average cost of the optimal solution by up to 30%. In addition, an analysis of drone payload sensitivity showed that drone payload capacity is negatively correlated with delivery time, and that larger customer sizes corresponded to higher sensitivity. Full article

Unsupervised anomalous sound detection (ASD) models the normal sounds of machinery through classification operations, thereby identifying anomalies by quantifying deviations. Most recent approaches adopt depthwise separable modules from MobileNetV2. Extensive studies demonstrate that squeeze-and-excitation (SE) modules can enhance model fitting by dynamically weighting input features to adjust output distributions. However, we observe that conventional SE modules fail to adapt to the complex spectral textures of audio data. To address this, we propose an Audio Texture Attention (ATA) specifically designed for machine noise data, improving model robustness. Additionally, we integrate an LSTM layer and refine the temporal feature extraction architecture to strengthen the model’s sensitivity to sequential noise patterns. Experimental results on the DCASE 2020 Challenge Task 2 dataset show that our method achieves state-of-the-art performance, with AUC, pAUC, and mAUC scores of 96.15%, 90.58%, and 90.63%, respectively. Full article

Online monitoring systems continuously collect massive multivariate time-series data from converter transformers. Accurate outlier detection in these data is essential for identifying sensor faults, communication errors, and incipient equipment failures, thereby ensuring reliable condition assessment and maintenance decisions. However, the complex characteristics of transformer monitoring data—including non-Gaussian distributions from diverse operational modes, high dimensionality, and multi-scale temporal dependencies—render traditional outlier detection methods ineffective. This paper proposes a Multi-View Clustering-based Outlier Detection (MVCOD) framework that addresses these challenges through complementary data representations. The framework constructs four complementary data views—raw-differential, multi-scale temporal, density-enhanced, and manifold representations—and applies four detection algorithms (K-means, HDBSCAN, OPTICS, and Isolation Forest) to each view. An adaptive fusion mechanism dynamically weights the 16 detection results based on quality and complementarity metrics. Extensive experiments on 800 kV converter transformer operational data demonstrate that MVCOD achieves a Silhouette Coefficient of 0.68 and an Outlier Separation Score of 0.81, representing 30.8% and 35.0% improvements over the best baseline method, respectively. The framework successfully identifies 10.08% of data points as outliers with feature-level localization capabilities. This work provides an effective and interpretable solution for ensuring data quality in converter transformer monitoring systems, with potential applications to other complex industrial time-series data. Full article

This paper addresses the core issues of slow coverage rate growth and high repeated detection rates in multi-UAV cooperative search operations within unknown areas. A distributed cooperative search algorithm based on the maximum entropy mechanism is proposed to resolve these challenges. It innovatively integrates the entropy gradient decision framework with DMPC-OODA (Distributed Model Predictive Control-Observe, Orient, Decide, Act) rolling optimization: environmental uncertainty is quantified through an exponential decay entropy model to drive UAVs to migrate toward high-entropy regions; element-wise product operations are employed to efficiently update environmental maps; and a dynamic weight function is designed to adaptively adjust the weights of coverage gain and entropy gain, thereby balancing “rapid coverage” and “accurate exploration”. Through multiple independent repeated experiments, the algorithm demonstrates significant improvements in coverage efficiency—by 6.95%, 12.22%, and 59.49%, respectively—compared with the Search Intent Interaction (SII) mode, non-entropy mode, and random mode, which effectively enhances resource utilization. Full article

Distribution network situational awareness prediction is a key technology for ensuring the safe and stable operation of distribution networks. However, most existing methods suffer from spatio-temporal dynamic correlation and dynamic topology, resulting in unsatisfactory performance. To address these issues, we propose a distribution network situational awareness prediction method based on a spatio-temporal attention dynamic graph neural network model that realizes the decoupling of spatio-temporal features of the distribution network data by adopting the alternating stacking of the multi-head self-attention mechanism with temporal dynamic perception and the spatial dynamic graph convolution module. Furthermore, the dynamic correlation matrix is introduced to adaptively adjust the node interaction weights to effectively handle the network dynamic topology information. Through extensive experiments, the proposed method outperforms eight baseline models. Full article

Generative video diffusion models (GVDs) generate high-fidelity, text-conditioned videos but risk producing unsafe or copyrighted content due to training on large, uncurated datasets. Concept erasure techniques aim to remove such harmful concepts from pre-trained models while preserving overall generative performance. However, existing methods mainly target single-concept erasure and thus cannot satisfy the demand for simultaneously eliminating multi-concept in real-world scenarios. On the one hand, naively applying single-concept erasure sequentially to multi-concept often yields suboptimal results due to conflicts among target concepts; on the other hand, methods that alter concept mappings exhibit very poor adaptability and fail to accommodate the dynamic concept changes. To address these, we propose ConceptVoid, a scalable multi-concept erasure framework formulated as a constrained multi-objective optimization problem. For each target concept, an erasure loss is defined as the discrepancy between noise predictions conditioned and unconditioned on the concept. Non-target generation capabilities are preserved via output-distribution alignment regularization. We apply the multiple gradient descent algorithm (MGDA) to obtain Pareto-optimal solutions, aiming to minimize conflicts among different concept erasure objectives. In addition, we improve MGDA by introducing an importance-weighting mechanism, which adjusts the weights of gradients corresponding to each erasure objective, enabling flexible control over the priority and intensity of erasing different concepts, thereby enhancing the scalability of ConceptVoid. Extensive experiments demonstrate the effectiveness of ConceptVoid, validating our key contributions: (1) a scalable framework for multi-concept erasure in GVDs; (2) the integration of per-concept erasure with distribution alignment to retain non-target quality; and (3) an enhanced MGDA for conflict-aware, controllable erasure. Full article

Unmanned Aerial Vehicles (UAVs) are known to have limited payloads, which challenges their widespread use in transporting heavy goods. Meanwhile, collaboration between multiple UAVs in performing such a task may be a promising solution. To address the issues associated with the simultaneous use of UAVs, this paper presents a formation control system for transporting a payload suspended via a cable using two UAVs. The control structure is based on a layered scheme that combines a null-space-based kinematic controller with a PID controller associated with each UAV (quadcopters) with a neural correction system. The null-space supervisor controller is designed to generate the desired velocity for the UAV system to maintain formation. This proposal aims to avoid obstacles, balance the weight distribution across each vehicle, and also reduce the payload trajectory tracking error. The PID controller associated with the neural correction system receives these desired speeds and performs dynamic compensation, taking into account parametric uncertainties and dynamic disturbances caused by the movement of the payload coupled to the UAV systems. The stability analysis of the entire control system is performed using Lyapunov theory. Detailed dynamic models of each UAV in the system, the flexible cables, and the payload are presented in a realistic scenario. Finally, numerical simulations demonstrate the good performance of the UAV system control in formation. Full article

Over the past 70 years, Central Asia has emerged as a globally recognized water security hotspot due to its unique geographic location and uneven distribution of water resources. In arid and semi-arid regions, understanding runoff dynamics under climate change is essential for ensuring regional water security. This study addresses the data-sparse Central Asian region by applying the ISIMIP3b multi-scenario analysis framework, selecting three representative global hydrological models. Using model intercomparison, trend analysis, and geographically weighted regression, we assess the spatiotemporal evolution of runoff from 1950 to 2080 and investigate the spatial heterogeneity of runoff responses to precipitation and temperature. The results show that under the historical scenario, all models consistently identify similar spatial pattern of runoff, with higher values in southeastern mountainous regions and lower values in western and central regions. However, substantial differences exist in runoff magnitude, with regional annual means of 10, 26, and 68 mm across the three models, respectively. The spatial disparity of runoff distribution is projected to increase under higher SSP scenarios. During the historical period, most of Central Asia experienced a slight decreasing trend in runoff, but the overall trends were −0.022, 0.1, and 0.065 mm/year, respectively. In contrast, future projections indicate a transition to increasing trends, particularly in eastern regions, where trend magnitudes and statistical significance are notably greater than in the west. Meanwhile, the spatial extent of significant trends expands under high-emission scenarios. Precipitation exerts a positive influence on runoff in over 80% of the region, while temperature impacts exhibit strong spatial variability. In the WaterGAP2-2e and MIROC-INTEG-LAND models, temperature has a positive effect on runoff in glaciated plateau regions, likely due to enhanced snow and glacier melt under warming conditions. This study presents a multi-model framework for characterizing climate–runoff interactions in data-scarce and environmentally sensitive regions, offering insights for water resource management in Central Asia. Full article

Synthetic Aperture Radar (SAR) plays a pivotal role in remote-sensing target detection. However, domain shift caused by distribution discrepancies across sensors, coupled with the scarcity of target-domain samples, severely restricts the generalization and practical performance of SAR detectors. To address these challenges, this paper proposes a few-shot SAR target-detection framework tailored for cross-sensor scenarios (CS-FSDet), enabling efficient transfer of source-domain knowledge to the target domain. First, to mitigate inter-domain feature-distribution mismatch, we introduce a Multi-scale Uncertainty-aware Bayesian Distribution Alignment (MUBDA) strategy. By modeling features as Gaussian distributions with uncertainty and performing dynamic weighting based on uncertainty, MUBDA achieves fine-grained distribution-level alignment of SAR features under different resolutions. Furthermore, we design an Adaptive Cross-domain Interactive Coordinate Attention (ACICA) module that computes cross-domain spatial-attention similarity and learns interaction weights adaptively, thereby suppressing domain-specific interference and enhancing the expressiveness of domain-shared target features. Extensive experiments on two cross-sensor few-shot detection tasks, HRSID→SSDD and SSDD→HRSID, demonstrate that the proposed method consistently surpasses state-of-the-art approaches in mean Average Precision (mAP) under 1-shot to 10-shot settings. Full article

Establishing scientifically sound and equitable compensation standards is crucial for effective ecological compensation. This study focuses on the quantitative assessment of ecological compensation standards in the water-source areas of the South-to-North Water Diversion Project. Based on the dual perspective of cost and benefit, we embed a three-dimensional dynamic adjustment coefficient—water volume, water quality, and payment capacity—and fully considered spillover effects. Using the AHP-Entropy Method, the allocation ratio of the water-receiving area was scientifically divided, achieving differentiated distribution and dynamic adaptation of the compensation mechanism. The compensation allocation ratio for water-receiving areas was determined, ensuring differentiated distribution and dynamic adaptability in the compensation mechanism. The results show the following: (1) In 2023, the ecological compensation amount for Yangzhou, based on the cost method and the equivalent factor method, ranges from CNY 1.21 billion to 2.53 billion. The amount of compensation after the dynamic game between both parties can avoid the waste of resources caused by over-compensation, and at the same time make up for the shortcomings of under-compensation due to the current water price. (2) Ecological compensation is measured only from the single perspective of the water-source area, without considering the differences in the receiving area. This paper uses the AHP-entropy value method to combine and assign weights, and calculates the apportionment ratio of the main water-receiving areas of Shandong Province in the east line of the South-to-North Water Diversion: for the Jiaodong line, these are Qingdao 20.97% and Jinan 14.53%, and for the North Shandong line, they are Dongying 23.98%, Dezhou 13.68%, Liaocheng 9.47%, and Binzhou 17.37%. (3) The dynamic correction coefficient and game model can effectively balance the cost of protecting the water-source area and the receiving area’s ability to pay, and combination with the empowerment method enhances the regional difference in suitability. Full article

Deep-learning-based multiple label chest X-ray classification has achieved significant success, but existing models still have three main issues: fixed-scale convolutions fail to capture both large and small lesions, standard pooling is lacking in the lack of attention to important regions, and linear classification lacks the capacity to model complex dependency between features. To circumvent these obstacles, we propose CONVFCMAE, a lightweight yet powerful framework that is built on a backbone that is partially frozen (77.08 % of the initial layers are fixed) in order to preserve complex, multi-scale features while decreasing the number of trainable parameters. Our architecture adds (1) an intelligent global pooling module that is learnable, with $1\times 1$ convolutions that are dynamically weighted by their spatial location, and (2) a multi-head attention block that is dedicated to channel re-calibration, along with (3) a two-layer MLP that has been enhanced with ReLU, batch normalization, and dropout. This module is used to enhance the non-linearity of the feature space. To further reduce the noise associated with labels and the imbalance in class distribution inherent to the NIH ChestXray14 dataset, we utilize a combined loss that combines BCEWithLogits and Focal Loss as well as extensive data augmentation. On ChestXray14, the average ROC–AUC of CONVFCMAE is 0.852, which is 3.97 percent greater than the state of the art. Ablation experiments demonstrate the individual and collective effectiveness of each component. Grad-CAM visualizations have a superior capacity to localize the pathological regions, and this increases the interpretability of the model. Overall, CONVFCMAE provides a practical, generalizable solution to the problem of extracting features from medical images in a practical manner. Full article

With the increasing content richness of social media platforms, Multimodal Named Entity Recognition (MNER) faces the dual challenges of heterogeneous feature fusion and accurate entity recognition. Aiming at the key problems of inconsistent distribution of textual and visual information, insufficient feature alignment and noise interference fusion, this paper proposes a multimodal named entity recognition model based on dual-stream Transformer: CASF-MNER, which designs cross-modal cross-attention based on visual and textual features, constructs a bidirectional interaction mechanism between single-layer features, forms a higher-order semantic correlation modeling, and realizes the cross relevance alignment of modal features; construct a dynamic perception mechanism of multimodal feature saliency features based on multiscale pooling method, construct an entropy weighting strategy of global feature distribution information to adaptively suppress noise redundancy and enhance key feature expression; establish a deep semantic fusion method based on hybrid isomorphic model, design a progressive cross-modal interaction structure, and combine with contrastive learning to realize global fusion of the deep semantic space and representational consistency optimization. The experimental results show that CASF-MNER achieves excellent performance on both Twitter-2015 and Twitter-2017 public datasets, which verifies the effectiveness and advancement of the method proposed in this paper. Full article