Pansharpening plays a critical role in remote sensing image processing by generating high-resolution multispectral (HRMS) images through the fusion of low-resolution multispectral (LRMS) and high-resolution panchromatic (PAN) images. In recent years, deep learning (DL)-based approaches have significantly advanced fusion performance via end-to-end modeling. However, existing methods still face notable limitations, including insufficient modeling of long-range dependencies, tight coupling between spectral and spatial information that may introduce spectral distortion, and high computational complexity that restricts practical deployment in high-resolution scenarios. To address these issues, we propose spectral-spatial-scale Mamba for pansharpening (S(3)Mamba-Pan), a frequency-aware state space framework that jointly models spectral structure, spatial texture, and multiscale context in a coordinated yet decoupled manner. The method first extracts a global spectral anchor (GSA) from the LRMS input to guide fusion, then performs wavelet-based frequency decoupling to separate low-frequency spectral components from high-frequency spatial details. A dual-stream Mamba architecture conducts parallel state space modeling for spectral correlation learning and PAN-guided texture enhancement, and an adaptive distribution recalibration (ADR) module aligns channel-wise statistics before reconstruction. Experiments on WorldView-3 (WV3), QuickBird (QB), and GaoFen-2 (GF2) demonstrate consistent improvements over representative convolutional neural network (CNN), transformer, and state-space model (SSM)-based baselines under reduced-resolution evaluation. Under full-resolution no-reference assessment on WV3, S(3)Mamba-Pan achieves D-lambda 0.0144 , D-s 0.0303 , and HQNR 0.960, indicating improved spectral fidelity and spatial consistency. Ablation and visualization analyses further corroborate the contribution of each component and provide interpretable evidence for the effectiveness of frequency decoupling and spectral anchoring. Code is available at https://github.com/FreeZS-a/S3Mamba

Isolated industrial microgrids impose extremely high requirements on the reliability of physical constraints and the response speed of real-time scheduling schemes. While traditional mixed-integer linear programming (MILP) can rigorously guarantee physical constraints, the computational latency caused by its NP-hard nature fails to meet the requirements of real-time management. Conversely, although existing end-to-end deep learning methods achieve rapid inference, they often face the risk of physical constraint violations. To this end, this article proposes an Edge-conditioned bipartite graph neural network (ECB-GNN), aiming to achieve efficient and robust branching decisions within the MILP solution framework. Through an edge-conditioned multihead attention mechanism, ECB-GNN models physical coefficients as edge features for bi-directional message passing, consistently enhancing the perception of strong coupling constraints in microgrids. Furthermore, we release a standardized microgrid MILP benchmark of 225 typical instances, incorporating dual-source perturbations and multilevel difficulty categorizations. Experimental results demonstrate that ECB-GNN consistently outperforms existing solvers and branching strategies in both accuracy and speed. It exhibits superior robustness against parametric fluctuations and topological changes, maintaining consistent acceleration even for large-scale instances with $10^{6}$ variables.

Rotating Synthetic Aperture (RSA) imaging technology has the advantage of lightweight design and no requirement for on-orbit assembly. It offers an innovative solution to overcome the physical limitations of traditional large-aperture optical systems for high-resolution, high-orbit Earth observation. However, the unique, anisotropic degradation inherent to its rectangular primary mirror poses a severe challenge for automated target interpretation. Existing object detection methods lack the perceptual and adaptive capabilities to handle this directional degradation, resulting in poor performance in the RSA system. To address this, we propose a novel end-to-end Degradation-Aware and Collaboratively Enhanced Detector (DACE-Det), tailored for object detection in anisotropically degraded images. The framework first employs the Latent Degradation Pattern Embedding (LDPE) module, based on contrastive learning, to perceive and disentangle the degradation prior from the image content. The prior then guides the Hierarchical Anisotropic Feature Learning Network (HAFNet), a backbone specifically designed to extract multi-level features from anisotropically degraded images. Concurrently, the parallel restoration branch generates a detail-rich feature stream that complements the backbone features. Furthermore, the Synergistic Expert Fusion Module (SEFM) leverages a Mixture-of-Experts mechanism to dynamically fuse detection and restoration feature streams. To support research in related fields, we construct RSA-Aircraft, the first object detection dataset with anisotropic degradation, based on a full-link imaging simulation model. The dataset contains 35,704 annotated degraded images of military and civilian airports from over 100 countries and regions worldwide, captured across various years, seasons, and weather conditions. Results from both digital simulations and semi-physical experiments demonstrate that DACE-Det significantly surpasses 27 mainstream methods and outperforms the baseline by 9.86 percentage points. It offers a practical solution for object detection in RSA imagery. The RSA-Aircraft dataset will be available at https://github.com/sgtojd/RSA-Aircraft-A-benchmark-Dataset-for-Object-Detection-with-Anisotropic-Degradation for research purposes.

Heavy haul railways, as a strategic and critical national infrastructure, constitute a core component of the comprehensive transportation system. Their safe and efficient operation is of vital importance to national economic stability and societal development. The Heavy haul train control system (HTCS), an innovative train control model in the field of heavy haul railways, plays a key role in ensuring the safety and reliability of train operations. This paper designs a multi-level controlled key management system based on a multi-level control (MultiC-KMS) framework, which to realize full life-cycle management of a three-tier key system. This study presents an in-depth analysis of various aspects, including system architecture, key generation and distribution strategies, update mechanisms, and secure encryption methods. Mathematical models for key life cycles and secure transmission are also developed. The results show that the proposed framework significantly enhances the communication security and controllability of the HTCS, thereby supporting the digital and intelligent development of railway systems.

The widespread integration of solar photovoltaic and wind generation has transformed the operating paradigm of modern power systems, and the resulting displacement of synchronous generators leaves hybrid microgrids facing low inertia, poor voltage regulation, and inaccurate reactive power sharing. Grid-forming (GFM) inverters using conventional droop control exhibit an inherent trade-off between voltage restoration and reactive power sharing due to mismatched feeder impedances. To address this gap, this paper proposes a distributed finite-time secondary control framework coupled with an adaptive virtual impedance mechanism for islanded hybrid AC microgrids. Using non-smooth feedback control and multi-agent consensus theory, the proposed method guarantees finite-time restoration of frequency and voltage to their nominal values, overcoming the slow convergence of asymptotic controllers. A distributed adaptive virtual impedance loop further compensates for line impedance mismatches in real time, ensuring exact proportional reactive power sharing without central coordination. The framework is validated using Lyapunov stability criteria and evaluated on a modified IEEE 34-node test system with multiple renewable and storage units. Results show a 42% reduction in convergence time and a 95% improvement in reactive power sharing accuracy relative to state-of-the-art baselines, even under severe communication delays and dynamic load variations.

The frequency regulation reserve setting of wind-PV-storage power stations is crucial. However, the existing grid codes set up the station reserve in a static manner, where the synchronous generator characteristics and frequency-step disturbance scenario are considered. Thus, the advantages of flexible regulation of renewable generations are wasted, resulting in excessive curtailment of wind and solar resources. In this study, a method for optimizing the frequency regulation reserve of wind PV storage power stations was developed. Moreover, a station frequency regulation model was constructed, considering the field dynamic response and the coupling between the station and system frequency dynamics. Furthermore, a method for the online evaluation of the station frequency regulation was proposed based on the benchmark governor fitting. This method helps in overcoming the capacity-based reserve static setting. Finally,an optimization model was developed, along with the proposal of the linearized solving algorithm. The field data from the JH4# station in China's MX power grid was considered for validation. The proposed method achieves a 24.77 % increase in the station income while ensuring the system frequency stability when compared with the grid code-based method.

Image registration is a prerequisite for many multi-source remote sensing image fusion applications. However, due to differences in imaging factors such as sensor type, imaging time, resolution, and viewing angle, multi-source image registration faces challenges of multidimensional coupling such as radiation, scale, and directional differences. To address this issue, this paper proposes a Siamese network based on cross-domain robust feature decoupling as an image registration framework (CRS-Net), aiming to improve the robustness of multi-source image features across domains, scales, and rotations. Firstly, we design Siamese multiscale encoders and introduce a rotation-invariant convolutional layer without additional training parameters, achieving natural invariance to any rotation. Secondly, we propose a modality-independent decoder that utilizes the self-similarity of feature neighborhoods to excavate stable high-order structural information. Thirdly, we introduce cluster-aware contrastive constraints to learn discriminative and stable keypoint pairs. Finally, we design three multi-source remote sensing datasets and conduct sufficient experiments. Numerous experimental results show that our proposed method outperforms other SOTA methods and achieves more accurate registration in complex multi-source remote sensing scenes.

At present, many infrared target detection approaches focus on designing modules that address the two key characteristics of targets: their weak signals and small size. However, these approaches often fail to fully leverage guided learning for weak and small target content, resulting in sub-optimal detection performance, particularly in terms of shape preservation and target positioning. To tackle this challenge, this paper proposes a multi-branch mutual-guiding learning network (MMLNet) that enhances the accuracy of infrared target detection, even in the absence of clear morphological and textural features in images. The method consists of three branches: edge, positioning, and detection, each of which is designed with a specialized module from a unique perspective. In the detection branch, we introduce a multi-dimensional lossless encoder optimized through a downsampling strategy and multi-level feature fusion to mitigate feature loss in small targets. In the positioning branch, a target positioning strategy is proposed to explicitly identify candidate targets from the image by means of a learnable multi-kernel pattern. In the edge branch, a simple architecture is adopted to enhance the ability of the model to preserve the target shape. To effectively utilize the knowledge of different branches, a mutual-guiding fusion module is developed to adjust information within and between branches. The manner adaptively utilizes the specific knowledge from each input branch. Experiment results demonstrate that the proposed method achieves comparable performance, and the visualization results show the advantages of our method in shape preservation and positioning of the targets. Our code is publicly available at https://github.com/qianngli/MMLNet.

Detecting aircraft in complex remote sensing scenes has significant value for military and civilian applications. To overcome the interference of complex environmental factors and achieve high-accuracy detection performance, the comprehensive utilization of optical and synthetic aperture radar (SAR) images for object detection has become a promising research direction. However, currently, there are problems with optical and SAR fusion detection, such as difficulty in obtaining paired registration training data and incomplete consideration of feature elements in the fusion model. To tackle these challenges, we present an aircraft detection method based on optical-SAR complementarity-aware feature fusion. First, an unpaired image translation model based on scattering feature enhancement GAN (SFEG) is designed to generate SAR images that are pixel-level registered with the input optical image. On this basis, a complementarity-aware feature fusion detection network (CFFDNet) combining differential feature spatial-aware complementary (DFSC) units and gate-generated weighted fusion (GWF) units is proposed to enhance the effective features of single-source image while improving the complementary fusion effect of multimodal features. Experiments on CORS-ADD and MAR20 datasets demonstrate that our method outperforms the compared classical single-modal and multimodal detection models. The latest code is available soon at: https://github.com/JimmyRSlab/Complementarity-aware-Feature-Fusion-for-Aircraft-Detection-via-Unpaired-Opt2SAR-Image-Translation

Remote sensing target detection technology in cloud and mist scenes is of great significance for applications such as marine safety monitoring and airport traffic management. However, the degradation and loss of features caused by the obstruction of cloud and mist elements still pose a challenging problem for this technology. To enhance object detection performance in adverse weather conditions, we propose a novel target detection method named CM-YOLO that integrates background suppression and semantic context mining, which can achieve accurate detection of targets under different cloud and mist conditions. Specifically, a component-decoupling-based background suppression (CDBS) module is proposed, which extracts cloud and mist components based on characteristic priors and effectively enhances the contrast between the target and the environmental background through a background subtraction strategy. Moreover, a local-global semantic joint mining (LGSJM) module is utilized, which combines convolutional neural networks (CNNs) and hierarchical selective attention to comprehensively mine global and local semantics, achieving target feature enhancement. Finally, the experimental results on multiple public datasets indicate that the proposed method realizes state-of-the-art performance compared to six advanced detectors, with mAP, precision, and recall indicators reaching 85.5%, 89.4%, and 77.9%, respectively.

Hyperspectral imaging (HSI) data pose both opportunities and challenges for target detection due to the high spectral resolution and vast data volume. Traditional band selection methods for HSI often prioritize image quality or information content, neglecting target distinctiveness in specific detection tasks. To address this issue, this work proposes a novel band selection method, genetic algorithm-based weighted constraint target band selection (GA-WCTBS), which utilizes an improved genetic algorithm to optimize band subsets for small target detection. GA-WCTBS prioritizes target distinctiveness and background clutter fluctuations by a proposed spectral signal-to-clutter ratio (SCR) inspired by the constraint target method, even in bands with lower image quality. It employs a genetic algorithm to consider the combinatorial potential of bands for optimal detection. Additionally, a k-means and weight assignment strategy improves the background estimation for selecting a band subset with better clutter suppression capability. Experiments on widely used public ABU and AVIRIS datasets demonstrate that the band subset selected by GA-WCTBS significantly outperforms the existing methods in terms of detection capability.

Infrared video satellites have the characteristics of wide-area long-duration surveillance, enabling continuous operation day and night compared to visible light imaging methods. Therefore, they are widely used for continuous monitoring and tracking of important targets. However, energy attenuation caused by long-distance radiation transmission reduces imaging contrast and leads to the loss of edge contours and texture details, posing significant challenges to target tracking algorithm design. This paper proposes an infrared small-target tracking method, the UIMM-Tracker, based on the tracking-by-detection (TbD) paradigm. First, detection uncertainty is measured and injected into the multi-model observation noise, transferring the distribution knowledge of the detection process to the tracking process. Second, a dynamic modulation mechanism is introduced into the Markov transition process of multi-model fusion, enabling the tracking model to autonomously adapt to targets with varying maneuvering states. Additionally, detection uncertainty is incorporated into the data association method, and a distance cost matrix between trajectories and detections is constructed based on scale and energy invariance assumptions, improving tracking accuracy. Finally, the proposed method achieves average performance scores of 68.5%, 45.6%, 56.2%, and 0.41 in IDF1, MOTA, HOTA, and precision metrics, respectively, across 20 challenging sequences, outperforming classical methods and demonstrating its effectiveness.

Aircraft detection and type identification in optical remote sensing imagery are critical for civilian and military applications, including air traffic control and strategic surveillance. However, existing methods ignore the unique cross-shaped geometric structure and low spatial occupancy of aircraft, leading to inaccurate localization and category confusion. In response, this paper proposes a novel anchor-free detection network that leverages point set representation, integrating the progressive class-aware dual branches (PCA-DB) and instance-guided enhancement module (IGEM). Specifically, considering the underlying structure of aircraft, PCA-DB consists of the coarse foreground instance branch and the refined cross-shaped branch to facilitate high-quality point set generation. Through multi-task learning, the auxiliary branches implicitly inject geometric priors into shared features, effectively suppressing background interference. Subsequently, IGEM introduces the interactive attention mechanism to adaptively fuse the instance-level information in the auxiliary branch with features in the main branches, explicitly enhancing the discriminative features of aircraft. Extensive experiments validate the superior performance of the proposed method on several aircraft datasets, including MAR20, FAIR1M-Plane, and CORS-ADD. There are 5.42%, 4.28%, and 1.37% improvements in mAP in our method compared to the baseline network.

Anomaly detection (AD) has emerged as a critical area of research in hyperspectral imagery (HIS) processing, focusing on detecting sparse, small targets with spectral and spatial features deviating from the background without prior information. The approach of AD based on reconstruction differences is a leading method in deep learning (DL) for hyperspectral AD (HAD). A key challenge is the accurate estimation of complex backgrounds. The essence of this challenge lies in accurately reconstructing background regions while inferring the latent background of anomaly regions. In this article, we propose a novel method called dual-window spectral diffusion (DWSDiff) for HAD. To address the challenge of complex background estimation in HSIs, we developed a spectral diffusion model specifically tailored for HSI. This model achieves precise background estimation through an iterative spectral diffusion and reverse reconstruction process. We also introduced a dual-window strategy to mitigate the influence of anomaly extension areas within the neighborhood on background estimation. Moreover, the scarcity of paired labeled HSIs from the same scene, with and without anomalies, limits the model's ability to learn features between anomaly and background. To address the shortage, we devised an anomaly generation strategy based on the principal component analysis (PCA) and the linear spectral mixing model (LSMM). Building on these, we designed a training and inference framework that integrates spectral diffusion, reverse background reconstruction, and target detection. Experimental results on the Airport-Beach-Urban (ABU) hyperspectral datasets demonstrate that DWSDiff outperforms 20 state-of-the-art (SOTA) HAD methods across six different areas under the curve (AUC) metrics.

Driven by the dual-carbon goals, the increasing penetration of distributed generations (DGs) enhances the flexibility and reliability of distribution networks. However, existing DG configuration studies mainly focus on ensuring the power supply to critical loads and improving system reliability, while often neglecting line vulnerability. To address this gap, this paper proposes a comprehensive line vulnerability assessment model integrating both structural and state perspectives. The line betweenness index and voltage stability index are fused using Dempster–Shafer evidence theory to accurately identify vulnerable lines. On this basis, a DG optimization configuration model considering line vulnerability is developed, which jointly optimizes DG siting and sizing. Simulation results verify that the proposed method effectively enhances system resilience and operational efficiency, achieving secure and optimal DG deployment.

In this article, a stator-PM counter-rotating dual-rotor high-speed flywheel electrical machine (SPMCRDR-HSEM) equipped with flux-intensifying permanent magnet (IPM) and split fault-tolerant teeth (SFTT) is proposed to improve the power density and mitigate the electromagnetic vibration. First, the topology of the SPMCRDR-HSEM, and its power density improvement principle, as well as the electromagnetic vibration mitigation principle are illustrated. Second, the theoretical analysis is carried out for investigating the influence of the IPM on the power density and that of the SFTT on the electromagnetic vibration. Then, a detail performance comparison analysis of SPMCRDR-HSEMs equipped with IPM and SFTT, IPM and conventional FTT, conventional PM and SFTT, respectively, is conducted. It shows that the SPMCRDR-HSEM with IPM and SFTT is superior to the other two machines in power density, torque quality, and electromagnetic vibration. Finally, a prototype of the SPMCRDR-HSEM equipped with IPM and SFTT is designed and fabricated, and the theoretical and finite element analysis are validated by experimental test.

The dynamic frequency control processes and economic operations of the large-scale power grids are separately applied. However, for the small inertia microgrids (MGs), the operating conditions tend to be more volatile due to relatively more uncontrollable entities being integrated. Hence, the frequency control solution of MGs should take the economic operation of MGs at a time-scale that is much shorter than the traditional economic dispatch of the large power grids. To this end, this paper proposes a two-layer coordinated frequency control strategy for MGs with enhanced economic operation consideration. For the upper optimal power-sharing layer, the distributed bisection algorithm is applied to obtain the optimal power sharing among heterogeneous resources. For the lower control layer, an autonomous control strategy that integrates both primary control and secondary control reference is applied by adopting the event-trigger mechanism. The proposed control approach can realize integrated active power control of MGs by simultaneously taking primary control, secondary control, and economic operation into consideration. Simulation studies with a heterogeneous resources-powered MG demonstrate its effectiveness.

Solid oxide electrolysis cell (SOEC) hydrogen production technology can range in size from small, appliance-size equipment to large-scale, central production facilities that can be tied directly to renewable or non-greenhouse-gas-emitting forms of electricity production, making it an ideal resource for demand response (DR). The SOEC hydrogen production system is a complex integrated system that encompasses fluid dynamics, electrical dynamics, and electrochemical and thermal dynamics, all of which involve non-linearity and non-convexity. Proper control of the SOEC hydrogen production system is crucial to enable its participation in the DR program. To overcome the difficulty of designing an explicit control law for such nonlinear systems with nonconvex optimization features in DR applications, deep reinforcement learning (DRL) is explored to achieve the optimal control of the SOEC system for DR participation. Specifically, a twin delayed deterministic policy gradient (TD3) control framework is applied to achieve optimal response performance during DR events by considering power tracking error and hydrogen production efficiency with a suitable reward function. Two case studies with grid connections for tracking different DR commands were investigated. The first case study involved operating conditions reaching the boundaries, while the second involved operating conditions within the boundaries. The results showed that the proposed DRL-based control for SOEC can track the DR signal in a timely manner while maintaining high energy efficiency

Aircraft detection and recognition in satellite imagery can provide crucial technical support for military and civilian applications such as battlefield early warning and air route control. However, existing methods struggle to meet the high accuracy demands due to the interference from complex scenes, as well as the high intra-class variations and low inter-class differences from the aircraft. To address the above challenges, this work fully exploits the potential of existing detectors and proposes a multi-model ensemble prediction framework for aircraft detection and recognition. Specifically, the complementary strengths of Transformer and CNN architectures are leveraged by incorporating multi-scale image inputs and decision-level fusion. The ensemble prediction mitigates the interference from complex scenes and assists the network in making more accurate decisions. On this basis, the oriented progressive prediction is proposed by cascading multiple prediction stages after the region proposal network for each model, which refines the detection and recognition results. Besides, diverse augmentation strategies are employed to generate various training samples, which enhances the model’s generalization capability. Experimental results prove that the proposed method can improve the mean Average Precision (mAP) by over 10% compared to single-model prediction.