Search Results (4,968)

Accurate and timely precipitation forecasting is critical for climate risk management, agriculture, and hydrological regulation. However, this task remains challenging due to the dynamic evolution of atmospheric systems, heterogeneous environmental factors, and frequent missing data in multi-source observations. To address these issues, we propose DAHG, a novel long-term precipitation forecasting framework based on dynamic augmented heterogeneous graphs with reinforced graph generation, contrastive representation learning, and long short-term memory (LSTM) networks. Specifically, DAHG constructs a temporal heterogeneous graph to model the complex interactions among multiple meteorological variables (e.g., precipitation, humidity, wind) and remote sensing indicators (e.g., NDVI). The forecasting task is formulated as a dynamic spatiotemporal regression problem, where predicting future precipitation values corresponds to inferring attributes of target nodes in the evolving graph sequence. To handle missing data, we present a reinforced dynamic graph generation module that leverages reinforcement learning to complete incomplete graph sequences, enhancing the consistency of long-range forecasting. Additionally, a self-supervised contrastive learning strategy is employed to extract robust representations of multi-view graph snapshots (i.e., temporally adjacent frames and stochastically augmented graph views). Finally, DAHG integrates temporal dependency through long short-term memory (LSTM) networks to capture the evolving precipitation patterns and outputs future precipitation estimations. Experimental evaluations on multiple real-world meteorological datasets show that DAHG reduces MAE by % and improves R² by .02 over state-of-the-art baselines (p < 0.01), confirming significant gains in accuracy and robustness, particularly in scenarios with partially missing observations (e.g., due to sensor outages or cloud-covered satellite readings). Full article

The Retrieval-Augmented Generation (RAG) framework shows great potential in terms of improving the reasoning and knowledge utilization capabilities of language models. However, most existing RAG systems heavily rely on large language models (LLMs) and suffer severe performance degradation when using small language models (SLMs), which limits their efficiency and deployment in resource-constrained environments. To address this challenge, we propose Retrieval-Adaptive-Planning RAG (RAP-RAG), a lightweight and high-efficiency RAG framework with adaptive retrieval task planning that is compatible with both SLMs and LLMs simultaneously. RAP-RAG is built on three key components: (1) a heterogeneous weighted graph index that integrates semantic similarity and structural connectivity; (2) a set of retrieval methods that balance efficiency and reasoning power; and (3) an adaptive planner that dynamically selects appropriate strategies based on query features. Experiments on the LiHua-World, MultiHop-RAG, and Hybrid-SQuAD datasets show that RAP-RAG consistently outperforms representative baseline models such as GraphRAG, LightRAG, and MiniRAG. Compared to lightweight baselines, RAP-RAG achieves 3–5% accuracy improvement while maintaining high efficiency and maintains comparable efficiency in both small and large model settings. In addition, our proposed framework reduces storage size by 15% compared to mainstream frameworks. Component analysis further confirms the necessity of weighted graphs and adaptive programming for robust retrieval under multi-hop reasoning and heterogeneous query conditions. These results demonstrate that RAP-RAG is a practical and efficient framework for retrieval-enhanced generation, suitable for large-scale and resource-constrained scenarios. Full article

In cross-corpus scenarios, inappropriate feature-processing methods tend to cause the loss of key emotional information. Additionally, deep neural networks contain substantial redundancy, which triggers domain shift issues and impairs the generalization ability of emotion recognition systems. To address these challenges, this study proposes a cross-corpus speech emotion recognition model based on attention-driven feature refinement and spatial reconstruction. Specifically, the proposed approach consists of three key components: first, an autoencoder integrated with a multi-head attention mechanism to enhance the model’s ability to focus on the emotional components of acoustic features during the feature compression process of the autoencoder network; second, a feature refinement and spatial reconstruction module designed to further improve the extraction of emotional features, with a gating mechanism employed to optimize the feature reconstruction process; finally, the Charbonnier loss function adopted as the loss metric during training to minimize the difference between features from the source domain and target domain, thereby enhancing the cross-domain robustness of the model. Experimental results demonstrated that the proposed method achieved an average recognition accuracy of 46.75% across six sets of cross-corpus experiments, representing an improvement of 4.17% to 14.33% compared with traditional domain adaptation methods. Full article

Digital mining has been evolving in recent years under the Industry 4.0 paradigm. In this sense, technological tools such as sensors aid the management and operation of mining projects, reducing the risk of accidents, increasing productivity, and promoting business sustainability. DT is a technological tool that enables the integration of various Industry 4.0 technologies to create a virtual model of a real, physical entity, allowing for the study and analysis of the model’s behavior through real-time data collection. A digital twin of an underground mine is a real-time, virtual replica of an actual mine. It is like an extremely detailed “simulator” that uses data from sensors, machines, and personnel to accurately reflect what is happening in the mine at that very moment. Some of the functionalities of an underground mining DT include (i) accurate geometry of the real physical asset, (ii) real-time monitoring capability, (iii) anomaly prediction capability, (iv) scenario simulation, (v) lifecycle management to reduce costs, and (vi) a support system for smart and proactive decision-making. A digital twin of an underground mine offers transformative benefits, such as real-time operational optimization, improved safety through risk simulation, strategic planning with predictive scenarios, and cost reduction through predictive maintenance. However, its implementation faces significant challenges, including the high technical complexity of integrating diverse data, the high initial cost, organizational resistance to change, a shortage of skilled personnel, and the lack of a comprehensive, multi-layered conceptual framework for an underground mine digital twin. To overcome these barriers and gaps, this paper proposes a strategy that includes defining an advanced, multi-layered conceptual framework for the digital twin. Simultaneously, it advocates for fostering a culture of change through continuous training, establishing partnerships with specialized experts, and investing in robust sensor and connectivity infrastructure to ensure reliable, real-time data flow that feeds the digital twin. Finally, validation of the advanced multi-layered conceptual framework for digital twins of underground mines is carried out through a questionnaire administered to a panel of experts. Full article

The precise prediction of high-speed railway bridge pier settlement plays a crucial role in construction, maintenance, and long-term operation; however, current mainstream prediction methods mostly rely on independent analyses based on traditional or hybrid models, neglecting the impact of geological and environmental factors on subsidence. To address this issue, this paper proposes a multi-factor settlement prediction model for high-speed railway bridge piers named the Reversible Instance Normalization Multi-Scale Adaptive Resolution Stream CMamba, abbreviated as RMCMamba. During the data preprocessing process, the Enhanced PS-InSAR technology is adopted to obtain the time series data of land settlement in the study region. Utilizing the cubic improved Hermite interpolation method to fill the missing values of monitoring and considering the environmental parameters such as groundwater level, temperature, precipitation, etc., a multi-factor high-speed railway bridge pier settlement dataset is constructed. RMCMamba fuses the reversible instance normalization (RevIN) and the multiresolution forecasting head (MARSHead), enhancing the model’s long-range dependence capture capability and solving the time series data distribution drift problem. Experimental results demonstrate that in the multi-factor prediction scenario, RMCMamba achieves an MAE of 0.049 mm and an RMSE of 0.077 mm; in the single-factor prediction scenario, the proposed method reduces errors compared to traditional prediction approaches and other deep learning-based methods, with MAE values improving by 4.8% and 4.4% over the suboptimal method in multi-factor and single-factor scenarios, respectively. Ablation experiments further verify the collaborative advantages of combining reversible instance normalization and the multi-resolution forecasting head, as RMCMamba’s MAE values improve by 5.8% and 4.4% compared to the original model in multi-factor and single-factor scenarios. Hence, the proposed method effectively enhances the prediction accuracy of high-speed railway bridge pier settlement, and the constructed multi-source data fusion framework, along with the model improvement strategy, provides technological and experiential references for relevant fields. Full article

The rapid advancement of 5G/6G networks and the Internet of Things has rendered mobile traffic patterns increasingly complex and dynamic, posing significant challenges to achieving precise cell-level traffic forecasting. Traditional deep learning models, such as LSTM and CNN, rely heavily on substantial datasets. When confronted with new base stations or scenarios with sparse data, they often exhibit insufficient generalisation capabilities due to overfitting and poor adaptability to heterogeneous traffic patterns. To overcome these limitations, this paper proposes a meta-learning framework—GMM-MCM-NF. This framework employs a Gaussian mixture model as a probabilistic meta-learner to capture the latent structure of traffic tasks in the frequency domain. It further introduces a multi-component synthesis mechanism for robust weight initialisation and a negative feedback mechanism for dynamic model correction, thereby significantly enhancing model performance in scenarios with small samples and non-stationary conditions. Extensive experiments on the Telecom Italia Milan dataset demonstrate that GMM-MCM-NF outperforms traditional methods and meta-learning baseline models in prediction accuracy, convergence speed, and generalisation capability. This framework exhibits substantial potential in practical applications such as energy-efficient base station management and resilient resource allocation, contributing to the advancement of mobile networks towards more sustainable and scalable operations. Full article

In high-density cities, integrating photovoltaic shading devices (PVSDs) with urban block typology optimization is crucial for low-carbon development, yet the understanding of their synergistic effects remains limited. This study develops a novel multi-scale evaluation framework that bridges block-building hierarchies to address this research gap. Through parametric modeling, this study coupled 27 representative office block morphologies with 18 PVSDs in Wuhan, a prototype city for China’s hot-summer–cold-winter climate zone, systematically generating 486 scenarios for comprehensive evaluation. Using Rhino–Grasshopper (7.0) with Ladybug (1.7), Honeybee (1.6), and EnergyPlus (9.4), we then examined urban block typology-PVSDs interactions across these scenarios. Our findings demonstrate that coordinated block typology and PVSD variables serve as critical determinants of energy-performance synergy. High-Rise Hybrid blocks emerge as the superior configuration for integrated performance, achieving maximal passive energy savings, optimal renewable energy utilization, and substantial carbon reduction. PVSDs that are 0.4 m in width, with specific distance-to-width ratios, yield the highest integrated benefits. This work advances sustainable urban design by establishing a morphology–energy nexus framework, providing architects and urban planners with actionable strategies for climate-responsive design in similar regions, with direct implications for maximizing energy–PV synergy through morphology-aware design approaches. Full article

To address the governance challenges of “delayed response, fragmented strategies, and cognitive disconnection” in traditional emergency management, this paper proposes an intelligent framework—Scenario–Learning–Decision (SLD)—powered by Large Language Models (LLMs). The framework integrates Multi-Agent Systems (MAS) and prospect theory-based parameter modeling to build an emergency simulation platform featuring scenario perception, human–AI learning, and collective decision-making. Using the 2022 wildfire in City C as a case study, the research verifies the effectiveness of the SLD model in complex emergency contexts and provides theoretical support and practical pathways for developing human-centered intelligent emergency decision-making systems. Full article

Urban mobility in Bangkok is constrained by congestion, modal fragmentation, and gaps in First and Last Mile (FLM) access. This study develops a GIS-based framework that combines maximal-coverage location allocation with post-optimization accessibility diagnostics to inform intermodal hub siting. The network model compares one-, three-, and five-hub configurations using a 20 min coverage standard, and we conduct sensitivity tests at 15 and 25 min to assess robustness. Cumulative isochrones and qualitative overlays on BTS, MRT, SRT, Airport Rail Link, and principal water routes are used to interpret spatial balance, peripheral reach, and multimodal alignment. In the one-hub scenario, the model selects Pathum Wan as the optimal central node. Transitioning to a small multi-hub network improves geographic balance and reduces reliance on the urban core. The three-hub arrangement strengthens north–south accessibility but leaves the west bank comparatively underserved. The five-hub configuration is the most spatially balanced and network-consistent option, bridging the west bank and reinforcing rail interchange corridors while aligning proposed hubs with existing high-capacity lines and waterway anchors. Methodologically, the contribution is a transparent workflow that pairs coverage-based optimization with isochrone interpretation; substantively, the findings support decentralized, polycentric hub development as a practical pathway to enhance FLM connectivity within Bangkok’s current network structure. Key limitations include reliance on resident population weights that exclude floating or temporary populations, use of typical network conditions for travel times, a finite pre-screened candidate set, and the absence of explicit route choice and land-use intensity in the present phase. Full article

In view of the special dispatching demands of isolated islands in low-density periods of renewable energy power generation, the defects of the traditional dispatching mode when applied to isolated power generation systems are analyzed, and the idea of reasonably extending the daily scheduling cycle is proposed to adapt to the application of flexible energy resources in the form of energy packages under various uncertain scenarios. Under the multi-party cooperative power supply strategy for isolated islands, we analyze the shortcomings of key element modeling. A global optimal model of energy scheduling for isolated islands considering low-density energy output periods is constructed based on a refined element model, and a corresponding solution is proposed for the nonlinear constraints. The reasonability and effectiveness of the refined model, the global optimal model, and the assumption of an extended scheduling cycle are verified by theoretical analysis and case simulation. Full article

The effectiveness of mixing ventilation for contaminant removal and maintaining indoor air quality remains an active topic of debate. In shared multi-person spaces, it is common for occupants to experience uneven exposure levels due to variations in system configuration and seating arrangements. Previous studies have primarily considered static occupancy schemes, leaving a gap in understanding how dynamic patterns of use interact with ventilation design. This study investigates the combined effects of system settings and occupancy patterns on ventilation effectiveness (VE), while also exploring whether lower ventilation rates can still sustain acceptable air quality. Validated Computational Fluid Dynamics (CFD) models were developed to simulate multiple scenarios involving three ceiling heights, two inlet and exhaust configurations, and three occupancy patterns. Analysis of air quality at the breathing zone reveals that the spatial arrangement of ventilation inlets and exhausts substantially influences VE, with optimized layouts improved system effectiveness by approximately 20%. Seating arrangement was similarly important, with favorable positioning relative to inlets improving perceived air quality by up to 25%. In addition, modest increases in ceiling height reduced the ventilation rate needed to maintain equivalent air quality, suggesting opportunities for energy savings without compromising occupant health. Overall, this study demonstrates that the interaction between system configuration and occupancy has a stronger impact on ventilation performance. These findings underscore the importance of integrated design strategies that align ventilation layout with occupant distribution to achieve both efficiency and equity in indoor environments. Full article

Simultaneous Localization and Mapping (SLAM) utilizes sensor data to concurrently construct environmental maps and estimate its own position, finding wide application in scenarios like robotic navigation and augmented reality. SLAM systems based on 3D Gaussian Splatting (3DGS) have garnered significant attention due to their real-time, high-fidelity rendering capabilities. However, in real-world environments containing dynamic objects, existing 3DGS-SLAM methods often suffer from mapping errors and tracking drift due to dynamic interference. To address this challenge, this paper proposes BDGS-SLAM—a Bayesian Dynamic Gaussian Splatting SLAM framework specifically designed for dynamic environments. During the tracking phase, the system integrates semantic detection results from YOLOv5 to build a dynamic prior probability model based on Bayesian filtering, enabling accurate identification of dynamic Gaussians. In the mapping phase, a multi-view probabilistic update mechanism is employed, which aggregates historical observation information from co-visible keyframes. By introducing an exponential decay factor to dynamically adjust weights, this mechanism effectively restores static Gaussians that were mistakenly culled. Furthermore, an adaptive dynamic Gaussian optimization strategy is proposed. This strategy applies penalizing constraints to suppress the negative impact of dynamic Gaussians on rendering while avoiding the erroneous removal of static Gaussians and ensuring the integrity of critical scene information. Experimental results demonstrate that, compared to baseline methods, BDGS-SLAM achieves comparable tracking accuracy while generating fewer artifacts in rendered results and realizing higher-fidelity scene reconstruction. Full article

Flood risk in small streams is rising under climate change, as small catchments are highly vulnerable to short, intense storms. We develop a high-resolution assessment that integrates a Digital Surface Model (DSM), a Digital Elevation Model (DEM), and airborne LiDAR within a MATLAB (2025b) hydraulic workflow. A hybrid elevation model uses the DEM as baseline and selectively retains DSM-derived structures (levees, bridges, embankments), while filtering vegetation via DSM–DEM differencing with a 1.0 m threshold and a 2-pixel kernel. We simulate 10-, 30-, 50-, 100-, and 200-year return periods and calibrate the 200-year case to the July 2025 Sancheong event (793.5 mm over 105 h; peak 100 mm h⁻¹). The hybrid approach improves predictions over DEM-only runs, capturing localized depth increases of 1.5–2.0 m behind embankments and reducing false positives in vegetated areas by 12–18% relative to raw DSM use. Multi-frequency maps show progressive expansion of inundation; in the 100-year scenario, 68% of the inundated area exceeds 2.0 m depth, while 0–1.0 m zones comprise only 13% of the footprint. Unlike previous DSM–DEM studies, this work introduces a selective integration approach that distinguishes structural and vegetative features to improve the physical realism of small-stream flood modeling. This transferable framework supports climate adaptation, emergency response planning, and sustainable watershed management in small-stream basins. Full article

This study takes the power battery pack of a pure electric vehicle as the research object, focusing on safety—a core concern widely emphasized in the automotive industry. In practical application scenarios, evaluating the safety of the power battery pack through a single operating condition fails to fully reflect its comprehensive safety performance throughout the vehicle’s entire life cycle. To overcome this limitation, a systematic analysis process was established. First, Catia geometric modeling software was used to simplify the battery pack structure, and HyperMesh was then employed for mesh generation. Second, three core analyses were conducted: static analysis, modal analysis, and extrusion condition analysis. A multi-condition safety evaluation system for electric vehicle battery packs during computer simulation analysis was proposed, which evaluates the battery pack from three dimensions: “dynamic stiffness-static strength-extrusion safety”. Results show that: modal analysis reveals the battery pack’s low-order natural frequencies exceed the vehicle’s excitation frequency (excitation point on the case cover); static analysis confirms it meets operational requirements; extrusion verification proves its safety complies with new national standards. The coupling effect of this multi-dimensional analysis breaks through the limitations of safety performance evaluation under a single operating condition, more realistically reflecting the battery pack’s comprehensive safety over its life cycle and providing a more systematic basis for power battery pack optimization. Full article

Time-sensitive networking (TSN), a cutting-edge technology enabling efficient real-time communication and control, provides strong support for traditional Ethernet in terms of real-time performance, reliability, and deterministic transmission. In TSN systems, although time-triggered (TT) flows enjoy deterministic delay guarantees, audio video bridging (AVB) and best effort (BE) traffic still share link bandwidth through statistical multiplexing, a process that remains nondeterministic. This competition in shared memory switches adversely affects data transmission performance. In this paper, a priority queue threshold control policy is proposed and analyzed for mixed-critical traffic in time-sensitive networks. The core of this policy is to set independent queues for different types of traffic in the shared memory queuing system. To prevent low-priority traffic from monopolizing the shared buffer, its entry into the queue is blocked when buffer usage exceeds a preset threshold. A two-dimensional Markov chain is introduced to accurately construct the system’s queuing model. Through detailed analysis of the queuing model, the truncated chain method is used to decompose the two-dimensional state space into solvable one-dimensional sub-problems, and the approximate solution of the system’s steady-state distribution is derived. Based on this, the blocking probability, average queue length, and average queuing delay of different priority queues are accurately calculated. Finally, according to the optimization goal of the overall blocking probability of the system, the optimal threshold value is determined to achieve better system performance. Numerical results show that this strategy can effectively allocate the shared buffer space in multi-priority traffic scenarios. Compared with the conventional schemes, the queue blocking probability is reduced by approximately 40% to 60%. Full article