Search Results (541)

Accurate identification of weak nodes is a prerequisite for online security assessment, preventive control, and resilience enhancement in modern power systems. However, conventional single-layer graph-learning models mainly emphasize local neighborhood aggregation and are insufficient for characterizing vulnerability propagation from equipment-level disturbance to regional congestion and system-level transfer constraints. This paper proposes a mechanism-aware hierarchical graph-learning framework for weak-node identification in complex interconnected power grids. We emphasize that attention, fusion, and gating operations are standard neural-network mechanisms and are not claimed as new generic deep-learning blocks. The contribution of this paper is the power-system-specific formulation: constructing an electrically meaningful local-supernode hierarchy, defining reproducible mechanism-based node and branch-vulnerability proxies, and interpreting weak-node rankings through node–line–corridor coupling evidence. In the validated implementation, a local graph convolutional encoder and a supernode/global graph convolutional encoder generate 32-dimensional local embeddings and 16-dimensional global embeddings, which are concatenated and decoded by a 48 → 24 → 1 multilayer perceptron to obtain node vulnerability scores. Experiments are conducted on reproducible IEEE benchmark data generated from pandapower standard systems, with representative comparisons on the IEEE 57-bus, 145-bus, and 300-bus systems and a detailed structural interpretation on the IEEE 145-bus case. The present results validate the ability of the implemented local–global hierarchical model to reproduce the proposed mechanism-based vulnerability proxy on representative small- and medium-scale benchmarks. Full article

Multiple-fault events can initiate overload propagation and cascading outages, resulting in severe load loss and reduced system resilience. Therefore, beyond conventional protection concepts based on the (n − 1) criterion, there is also a need to address multiple-fault events to minimize loss of load. This paper presents an optimized overload tripping scheme to mitigate cascading outages in high-voltage grids under multiple-fault conditions, where selected line switches or circuit breakers are opened in a controlled manner to isolate limited grid sections, minimize interrupted load, and prevent further overload propagation. The method combines inverse definite minimum time relay modeling with a heuristic graph-search algorithm implemented in pandapower to identify feasible switching actions that minimize load loss while preventing overload propagation. The approach is demonstrated on SimBench high-voltage urban and mixed benchmark grids under double-line fault scenarios. In the urban grid, the proposed scheme reduces the maximum load loss from 34.0% to 2.4%, while in the mixed grid, the reduction is from 50.3% to 5.2%. A SAIFI-inspired resilience proxy is introduced to quantify the reduction in customer/load interruptions, showing a resilience improvement factor of about 3.6 for cascading scenarios. In addition, thermal inertia analysis indicates that corrective switching must be completed within approximately 5 min to remain within line-temperature limits. The analysis is based on quasi-steady-state power-flow and relay simulations; transient stability effects are outside the scope of this study. The results demonstrate that the optimized overload tripping scheme is a promising adaptive protection strategy for improving grid resilience under severe contingency conditions. Full article

Active voltage control (AVC) in active distribution networks coordinates the reactive power outputs of distributed inverters to maintain bus voltages within secure limits. Although multi-agent reinforcement learning (MARL) shows promise for AVC, current methods face three main limitations: graph topologies rely on unweighted adjacency, ignoring physical parameters like line impedance and electrical distance; centralized critics output a single global Q-value, leading to coarse spatial credit assignment; and temporal critic modules suffer from vanishing gradients and representation drift. To address these issues, we propose physics-informed spatiotemporal multi-agent value learning (PST-MA), a physics-informed spatiotemporal value-learning framework integrating three coupled designs: a physics-informed graph attention mechanism with electrical-distance-aware sparsification; node-conditional value outputs utilizing a replicated-graph diagonal-extraction strategy; and a temporal latent compression module featuring a gated bypass and late action fusion. Experiments on the IEEE 33-bus and 141-bus systems validate the effectiveness of the proposed PST-MA method. Results demonstrate that it consistently achieves a higher controllable ratio than baseline methods for coordinated voltage regulation under uncertainty. Full article

This study proposes a novel hybrid prediction model (QGCN-LSTM) that combines Quantum Graph Convolutional Networks (QGCN) with classical Long Short-Term Memory (LSTM). The model takes space-time data as input and employs a hierarchical graph-based quantum encoding strategy. Specifically, classical spatial features are first aggregated into critical regional hubs, which are then mapped into the Hilbert space through a dense quantum encoding layer. Multi-scale features are extracted through the collaborative computation of QGCN and quantum gated recurrent units, and a quantum attention module is introduced to dynamically screen key information. Finally, the prediction results are generated through quantum measurement and a classical output layer. In the space-time data prediction task of urban traffic flow, a benchmark model system covering classical, cutting-edge, and traditional architectures was constructed. The experimental results show that QGCN-LSTM utilizes quantum entanglement gates to establish non-local road network associations, dynamically allocate feature weights to enhance the impact of critical time steps, and achieves deep compression of lines through quantum line pruning technology, effectively alleviating the common problem of “poor plateau” in quantum neural network training. In terms of prediction accuracy, the mean absolute error (MAE) of its key hub nodes is reduced by 34.1% compared to the graph convolution LSTM (GCN-LSTM) model, and the Spatial Correlation Index (SCI) is improved to 0.89. In addition, it also shows excellent performance in dynamic response, edge computing efficiency, and other aspects, meeting the real-time requirements of the traffic signal control system. This study provides an effective paradigm for the application of quantum collaborative architecture in complex spatiotemporal prediction tasks. Full article

Along with the enhancement of people’s public health consciousness and the requirement for individual diet arrangement getting more urgent, the meal recommendation method, which is based on artificial intelligence, has hence become an active research domain in the field of intelligent health. One system that makes practical recommendations must deal with the user’s unclear queries, while at the same time, it must satisfy strict nutrient demands. A great number of existing methods at present either do not take into account verifiable food composition data, or they handle implicit dietary restrictions in a not good way. For solving these problems, we put forward CARE (Constraint-Aware Recipe Engine). Beginning from a mixed Retrieval-Augmented Generation (RAG) basic model (CARE v1.0), we have developed CARE v2.0, which is a suggestion engine that unites intention polish, knowledge graph enlargement, and rule-based checking in a unified working line. Instead of depending on huge black-box models, our framework utilizes an effective language model that possesses 1.5 B parameters. User inquiry content are undergone parsing to become structured nutrition targets; a food knowledge graph links abstract health notions to specific cooking materials; and the obtained candidate results are filtered in accordance with strict diet restrictions, with optional checking carried out by an automatic agent-based reviewer. Under a zero-shot cold-start situation, the system attains a semantic recall@5 of 0.825 on 400 k recipes coming from Recipe1M+ and a newly created fuzzy-query benchmark (CAREBench-150), and it thus has a better performance than dense retrieval baselines (0.550) as well as direct zero-shot prompting. The constraint satisfaction rate is located at 85.0% in fast mode, and it rises to 98.5% when the verification module is in the working state; therefore, it supports the safety of recommendations. These findings indicate that structured food knowledge, which matches a compact algorithmic framework, can therefore connect unclear user intentions and accurate nutrition requirements effectively. Full article

Indoor localization for unmanned aerial vehicles (UAVs) remains challenging in GNSS-denied environments due to the difficulty of position calibration of multiple ultra-wideband (UWB) anchors and the asynchronous fusion of heterogeneous sensors. This paper proposes a multi-sensor fusion localization framework that integrates multi-anchor one-shot calibration with factor graph optimization (FGO). First, Landmark Multidimensional Scaling (LMDS) is used to reconstruct the relative geometry of the anchors and the onboard tag from ranging measurements. Then, rigid Procrustes alignment is performed using a small number of anchors with known coordinates in the East–North–Up (ENU) frame to recover the transformation to the ENU frame, thereby enabling efficient position calibration of multiple UWB anchors and UAV pose initialization. Subsequently, a tightly coupled factor graph is constructed by incorporating inertial measurement unit (IMU) pre-integration, UWB ranging, laser rangefinder height measurements, and visual–inertial odometry (VIO) pose constraints. The resulting nonlinear optimization problem is solved using incremental smoothing, which improves robustness against non-line-of-sight (NLOS) errors and long-term drift. Experimental results on anchor calibration, public datasets, and real-world indoor UAV flights demonstrate that the proposed method improves the accuracy and robustness of indoor UAV localization. In particular, on the real-world rectangle trajectory, FGO-TC reduces the RMSE by approximately 38.8% compared with FGO-LC. Full article

Software-defined networking (SDN) provides a flexible solution to manage complex networks on demand by centralized control and programmability. However, efficiently optimizing network configurations to achieve load balance and improve service quality remains challenging. In this paper, we propose a novel SDN network state awareness and dynamic routing optimization method, termed DGSFN-DR. Hereby, we leverage a Graph Attention Network (GAT) to model the spatial dependencies of the network topology for its link graph. Then, we employ a Recurrent Neural Network (RNN) to capture the temporal dependencies of link states, including the lagged temporal features induced by routing algorithms, to improve the prediction accuracy of future link states. Our algorithm dynamically adjusts routing strategies to optimize network performance according to the predicted link weights with the dual graph spatiotemporal fusion prediction network (DG-SFN). Experimental results demonstrate that our DGSFN-DR outperforms other methods in various network traffic intensities and topologies. Specifically, it achieves improvements of $4\%$ to $15\%$ in latency, jitter, packet loss, and available bandwidth. In particular, the DGSFN-DR exhibits superior adaptability and optimization potential under high traffic loads and complex network topologies. This work expands dynamic routing optimization theory for SDN and new insights for practical network management. Full article

In distribution networks, traditional fault location methods have insufficient anti-interference capability and low accuracy in locating high-resistance grounding faults. To address these issues, a distribution network fault location method on the basis of SimAM-GraphSAGE-GAT is proposed. Firstly, the distribution network topology structure is converted into an adjacency matrix, and the electrical parameters of the faulty line are incorporated as node features into the graph structure of the network. Subsequently, the sampling and aggregation mechanism of GraphSAGE is used for learning node representation. Features are refined using SimAM. As a parameter-free attention mechanism, SimAM improves the ability of the model to capture important fault information. Then, the multi-head attention mechanism of GAT is introduced to enhance the representation of neighborhood relationships. Finally, GraphSAGE is utilized once again for deep aggregation, with a view to localizing faults by node classification. An IEEE 33-node distribution network model is adopted to verify the effectiveness of the algorithm in the experiment. The results show that this method can maintain high positioning accuracy even under the tested conditions, such as high-resistance grounding, noise interference, and data loss. Full article

High-penetration renewable energy significantly increases uncertainty, dynamic network coupling, and the need for secure and coordinated smart-grid dispatch. To address the limitations of conventional forecasting-based and static graph-based methods, this paper proposes a unified dispatch framework that integrates topology-informed dynamic graph learning, privacy-aware multi-agent symbiotic reinforcement learning, and structural causal intervention analysis. The dispatch problem is formulated as a constrained partially observable stochastic game, in which multiple agents coordinate generation adjustment, reserve allocation, and congestion-aware corrective actions under engineering constraints. A physics-informed dynamic graph convolutional module captures both fixed physical topology and stress-dependent operational couplings, while a KL-regularized multi-agent reinforcement learning scheme improves cooperative task allocation under renewable fluctuations. Federated optimization with Rényi differential privacy is introduced to protect sensitive local operational information during training. In addition, a structural causal module provides intervention-based interpretation of how wind variation, load escalation, and line stress affect dispatch cost, congestion risk, and renewable curtailment. Experiments on a public-trace-driven benchmark based on a modified IEEE 30-bus system show that the proposed method achieves the best overall performance among the compared baselines, reducing dispatch-cost RMSE to 3.82, locational-price MAE to 2.95, renewable curtailment to 4.8%, and the constraint-violation rate to 0.30%. Overall, the framework shows favorable performance on the test benchmark, provides post hoc intervention-based interpretation of dispatch outcomes, and is evaluated under a reproducible benchmark construction and assessment protocol. Full article

While vision-language models (VLMs) have been widely applied in zero-shot anomaly detection (ZSAD), their performance remains limited by the inability to distinguish fine-grained normal and abnormal textures, coupled with inadequate capabilities in detecting complex morphological anomalies. To address these limitations, this paper proposes BAG-CLIP (Bifurcated Attention Graph-Enhanced CLIP), a dual-path graph-enhanced zero-shot anomaly detection method. This approach employs a Bifurcated Self-Attention (BSA) module to decouple visual features, processing global semantics and spatial details separately to mitigate the inherent conflict between abstract semantic representation and precise spatial localization. A Self-Attention Graph (SAG) module is designed to model the topological structure of complex morphological anomalies. This module dynamically constructs visual features’ topological relationships and utilizes graph convolutions to aggregate neighborhood information, thereby enhancing the model’s representational capacity for diverse and complex morphological anomalies. Extensive experiments are conducted on five diverse industrial datasets, featuring complex transmission line backgrounds alongside general industrial scenarios. The proposed method is comprehensively evaluated against 11 state-of-the-art (SOTA) methods. On the EPED (Electrical Power Equipment Dataset) and MPDD datasets, BAG-CLIP outperforms the second-best methods in image-level AUROC (Area Under the Receiver Operating Characteristic Curve) by 3.7% and 2.8%, respectively. BAG-CLIP achieves superior performance in both zero-shot anomaly detection and segmentation. Full article

This paper examines the sixth-order generalized Boussinesq equation, which describes the dynamics of shallow-water waves, including the effects of surface tension. The study introduces Kudryashov’s R-function neural network approach for the first time, aiming to provide exact solutions to the nonlinear differential equation that represents the mathematical model of the sixth-order generalized Boussinesq equation. This technique incorporates the solutions of a nonlinear differential equation into neural networks, using them as an activation function within the hidden layer. In line with previous research on this topic, two categories of solutions are derived: solitary wave and shock wave solutions. Additionally, this paper includes 3D and 2D graphs to visually represent the solutions obtained. Full article

High-precision and reliable self-localization is essential for autonomous navigation systems. However, in urban canyons (urban environments with clusters of high-rise buildings), Global Navigation Satellite Systems (GNSS) suffer from severe multipath and Non-Line-of-Sight (NLOS) signal reception. This causes a theoretically unbounded positive bias in pseudorange measurements, significantly degrading positioning integrity. To address this challenge, this study proposes a novel GNSS/Inertial Measurement Unit (IMU) tightly coupled integrated navigation system using factor graph optimization (FGO) integrated with machine learning-based NLOS detection. To train the NLOS detection model, we utilized a dual-polarized antenna to label signals based on the strength difference between RHCP and LHCP components, achieving a detection accuracy of 0.89. A random forest classifier identifies NLOS signals, and based on its classification labels, the variance of the corresponding GNSS pseudorange factors within the FGO framework is dynamically inflated. This effectively mitigates the impact of outliers while preserving the graph topology. Experimental evaluations in dense urban environments demonstrated that the proposed method improves horizontal positioning accuracy by 84.8% compared to conventional standalone GNSS positioning. The dynamic integration of machine learning-based signal classification and tightly coupled FGO provides an extremely robust positioning solution, proven to meet the stringent reliability requirements demanded of autonomous systems even under severe signal obscuration. Full article

Dam deformation monitoring is essential for ensuring the safe operation of hydraulic structures, yet practical data are often compromised by missing values and noise, while spatial coupling among monitoring points further complicates prediction. To address these challenges, this study proposes a Spatio-Temporal Multi-Graph Attention Fusion Network (STMGAFN) for dam deformation prediction and risk early warning under incomplete data conditions. Data quality is enhanced through a data-quality-aware hierarchical adaptive imputation mechanism combined with a VMD–wavelet joint denoising strategy. A multi-graph spatial modeling framework integrating temporal similarity, spatial proximity, structural zoning, and measuring-line connectivity is constructed, and fuses multi-source spatial features through a lightweight adaptive attention mechanism. A parameter-sharing recursive probabilistic temporal modeling approach is adopted to jointly predict deformation values and their associated uncertainties. Based on the predicted confidence intervals, a four-level risk classification and early-warning scheme is established. Experimental results on real GNSS monitoring data from dam sites demonstrate that the proposed method achieves an RMSE of 0.3588 mm, an MAE of 0.1738 mm, and an R² of 0.9865, outperforming baseline models including LSTM, TCN, CNN-LSTM, and STGCN. Moreover, the correlation between predictive uncertainty and actual error reaches 0.892, verifying the effectiveness and reliability of the proposed method for dam safety monitoring under complex conditions. Full article

In contemporary museum design, architects increasingly treat spatial experience as a medium of visitor engagement, yet movement is often reduced to a problem of routing and orientation rather than recognised as engagement in its own right. This study shows how Steven Holl’s parallax operates as a motivational mechanism at the Kiasma Museum of Contemporary Art. Parallax, a phenomenological and ecological construct, is examined through oblique thresholds, overlapping perspectives, and layered illumination. Integrating phenomenology, ecological psychology, and spatial configuration analysis, this study links embodied perception to measurable spatial properties. Spatial relations were quantified using space syntax—axial line analysis, justified graphs, and isovist analysis—alongside luminance and visual saliency mapping of Kiasma’s second and third floors. The results reveal a dominant ring structure in which visibility tightens at thresholds and views shift continuously along the route. Pronounced brightness gradients accompany these transitions and intensify perceived change along the sequence. These coupled spatial and luminous strategies may encourage exploratory navigation, positioning wayfinding as integral to the museum experience. This study argues that parallax links spatial configuration to embodied engagement, emerging as a perceptual effect produced through the interaction of spatial layout, luminous modulation, and bodily movement rather than functioning as a fixed design principle. Full article

To address the issues of load aggregation and traffic congestion in faulty electricity–transportation coupled networks (ETCNs), this paper proposes an electric vehicle (EV) charging guidance strategy based on Graph Reinforcement Learning (GRL). First, a graph-structured feature extraction model is developed. The GraphSAGE module is employed to capture the multi-scale spatiotemporal features of the ETCN. The topological changes and energy-information interaction characteristics under fault scenarios are analyzed. Second, a Finite Markov Decision Process (FMDP) framework is established to address the stochastic and dynamic nature of EV charging behavior. The charging station selection and route planning problem is transformed into an agent decision-making process. A reward function is designed by incorporating voltage constraints, traffic flow constraints, and state-of-charge margin penalties. This ensures a balanced consideration of power grid security and traffic efficiency. The FMDP model is then solved using a Deep Q-Network (DQN) to achieve optimal EV charging guidance under fault conditions. Finally, case studies are conducted on a coupled simulation scenario consisting of an IEEE 33-node power distribution system and a 23-node transportation network. Results show that the proposed method reduces the system operation cost to 218,000 CNY, controls the voltage deviation rate of the distribution network at 3.1% in line with the operation standard, and enables the model to achieve stable convergence after only 250 training episodes. It can effectively optimize the charging load distribution and maintain the voltage stability of the power grid under fault conditions. Full article