Search Results (1,185)

The study presents a comprehensive spatio-temporal assessment of meteorological drought in the Upper Vistula basin, a region located in southern Poland. The analysis was based on monthly precipitation data from 30 meteorological stations covering the period 1961–2022. These data were used to calculate the Standardized Precipitation Index (SPI) for accumulation periods of 3, 6, 9, 12, 24, and 48 months. Drought events were identified using run theory, adopting a threshold of SPI < −1 for all accumulation periods. On this basis, drought characteristics were determined, including the number of identified drought episodes (N), average drought duration (ADD), average drought severity (ADS), and average drought intensity (ADI). The multi-scale analysis revealed a clear dependence of drought characteristics on the time scale. Short-term droughts (SPI-3 and SPI-6) occurred frequently and were characterized by high monthly intensity but short duration. In contrast, long-term droughts (SPI-24 and SPI-48) occurred less frequently, but were marked by much longer duration and greater cumulative severity, despite lower average intensity. Spatial analyses showed substantial heterogeneity of drought characteristics within the Upper Vistula basin. The western and south-western parts of the region were particularly exposed to frequent short-term droughts, whereas long-term droughts were less frequent, but more regional in nature and resulted from accumulated, multi-year precipitation deficits affecting groundwater resources and catchment retention. The presented findings provide valuable information for improving drought monitoring systems and adaptation strategies in the Upper Vistula basin and in other climatically diverse regions of Central Europe. Full article

Wildfires are a major natural hazard causing extensive ecological damage and endangering human survival. Previous studies on wildfires in China have mostly focused on specific regions or individual drivers, with limited systematic assessments at the long-term and national scales. The spatiotemporal patterns of wildfires and their multiple driving mechanisms under China’s diverse climatic regimes remain insufficiently understood. To bridge this gap, we combined MCD64A1 burned area data (2001–2023) with multi-source natural (meteorological, vegetation, and topographic) and anthropogenic factors, using random forest models at both the national and regional scales to examine the spatiotemporal patterns, dominant drivers, and response mechanisms of wildfires in China. The results revealed that: (1) Spatially, wildfires were concentrated in northeastern and southern China, which accounted for 86.20% of the total burned area. Temporally, northern wildfires were primarily a spring-dominated fire regime, with peak activity in March and April, whereas southern wildfires were winter-dominated, peaking in February. (2) At the national scale, elevation was the key topographic factor influencing wildfire occurrence (relative importance = 0.49), with low-elevation and gentle-slope areas being more fire-prone. At the regional scale, the driving factors exhibit spatial differentiation, forming a spatial pattern of topography-dominated and climate-dominated. (3) Partial dependence plot analysis revealed nonlinear and threshold responses. Fire probability increases rapidly when the soil moisture is below 20 mm, while extremely high land surface temperatures in arid regions suppress fire occurrence due to fuel limitations. This study enhances the understanding of spatially heterogeneous wildfire drivers in China and provides a scientific basis for region-specific wildfire prevention and management strategies. Full article

Addressing the challenge of coordinated dispatch between wind/solar and thermal power in new energy grids, this research proposes a thermal power unit output prediction method based on a Transformer-BiLSTM hybrid deep learning model. First, a simulated annealing algorithm optimizes the output configuration of solar thermal power plants to mitigate fluctuations in wind and solar combined generation. An ant colony-greedy algorithm is then integrated to determine the optimal dispatch data for thermal power units, constructing a high-quality training dataset under physical constraints. In the model design, a bidirectional long short-term memory network captures short-term temporal features, while the Transformer’s multi-head self-attention mechanism models long-term dependencies. The model innovatively incorporates the learnable positional encoding to enhance temporal awareness. Experimental results demonstrate accurate predictions, with the power constraint mechanism effectively correcting over-limit forecasts. This ensures 98.7% of predictions during low-load periods comply with unit technical specifications. Compared to existing methods, this model avoids data limitations and manual feature engineering bottlenecks through the end-to-end wind–solar–thermal mapping, providing a high-precision solution for dispatch decisions in renewable-dominated grids. Full article

Precise assessment of battery state of health (SoH) is vital for certifying consistent performance in order to enable maintenance of energy storage system. This work compares different deep learning methods to learn and predict the complex and nonlinear dynamics of battery. The models are developed and tested for predicting SoH using sequential degradation data from batteries. The effectiveness of these models is assessed using matrices such as RMSE, MAE and R², along with qualitative analysis. The experiment results show that the BiLSTM model performs better than the others. It has the lowest RMSE (0.90), the lowest MAE (0.72), and the highest R² (0.99), which highlights its enhanced ability to capture long-term temporal dependencies. The proposed models are validated using NASA lithium-ion battery aging dataset (B0005), which is widely used as a benchmark for battery health predictions studies. Overall, the findings indicate that bidirectional network architecture significantly improves the accuracy and consistency of SoH predictions when compared to unidirectional models. Full article

Accurate and physically consistent state prediction is essential for shipboard power systems (SPS) operating under dynamic conditions. However, purely data-driven models often exhibit degraded robustness and physically inconsistent outputs when exposed to transient disturbances or limited data coverage. To address these limitations, this paper proposes a physics-constrained hybrid prediction model that integrates a convolutional neural network–bidirectional long short-term memory (CNN–BiLSTM) architecture with wide residual connections (WRC) and a physics-constrained loss (PCL). The proposed modeling approach combines real operational measurement data with high-resolution simulation data to enhance data diversity and improve generalization capability. The CNN–BiLSTM structure captures nonlinear temporal dependencies, while the WRC preserves critical low-level transient electrical features during deep temporal modeling. In addition, multiple physical constraints, including power balance, voltage conversion relationships, and battery state-of-charge (SOC) dynamics, are incorporated into the training process to enforce physically consistent predictions. The model is validated using charging and discharging experiments on a laboratory-scale SPS under both steady-state and transient conditions. Comparative results demonstrate that the proposed approach achieves higher prediction accuracy, improved dynamic stability, and faster recovery following disturbances compared with conventional data-driven models. These results indicate that physics-constrained deep learning provides an effective and interpretable modeling framework for SPS state prediction, supporting digital twin-oriented monitoring and real-time prediction applications. Full article

Dynamic Spectrum Access (DSA) has emerged as a viable solution to address spectrum scarcity in shared-spectrum networks. In response, the FCC established the Citizens Broadband Radio Service (CBRS) to manage and facilitate shared use of the federal and non-federal spectrum in a three-tiered access and authorization framework. However, due to the open nature of spectrum access and the usually limited coverage of the monitoring infrastructure, enforcing access rights in a shared-spectrum network becomes a daunting challenge. In this paper, we stipulate the use of crowdsourcing as a viable approach to engaging volunteers in spectrum monitoring in order to enforce spectrum access rights robustly and reliably. The success of this approach, however, hinges strongly on ensuring that spectrum access enforcement is carried out by reliable and trustworthy volunteers within the monitored area. To this end, a hybrid spectrum monitoring framework is proposed, which relies on opportunistically recruiting volunteers to augment the otherwise limited infrastructure of trusted devices. Although a volunteer’s participation has the potential to enhance monitoring significantly, their mobility may become problematic in ensuring reliable coverage of the monitored spectrum area. To ensure continued monitoring, inspite of volunteer mobility, deep learning-based models are used to predict the likelihood that a volunteer will be available within the monitoring area. Three models, namely LSTM, GRU, and Transformer, are explored to assess their feasibility and viability to predict a volunteer’s availability likelihood over an extended time interval, in a given spectrum monitoring area. Recurrent Neural Networks (RNNs) such as GRU and LSTM are effective for tasks involving sequential data, where both spatial and temporal patterns matter, which is the focus of volunteer availability prediction in spectrum monitoring. Transformers, on the other hand, excel at handling long range dependencies and contextual understanding. Furthermore, their parallel processing capabilities allows faster training and inference compared to RNN-based models like GRU and LSTM. A simulation-based study is developed to assess the performance of these models, and carry out a comparative analysis of their ability to predict volunteers’ availability to monitor the spectrum reliably. To this end, a real-world trace dataset of volunteers’ location, collected over five years, is used. The simulation results show that the three models achieve high prediction accuracy of volunteers’ availability, ranging from 0.82 to 0.92. The results also show that a GRU-based model outperforms LSTM and Transformer-based models, in terms of accuracy, Root Mean Square Error (RMSE), geodesic distance, and execution time. Full article

(1) Background: The scarcity and imbalance of real fault data significantly limit the development of data-driven fault diagnosis methods for liquid rocket engines (LREs), especially under few-shot conditions. (2) Methods: To address this issue, this study proposes a GAN-based fault data augmentation framework for multivariate LRE time-series signals and a hybrid diagnostic classifier combining convolutional neural networks (CNNs), bidirectional long short-term memory (BiLSTM), and multi-head attention (MHA). The GAN component is introduced to alleviate fault-data scarcity and class imbalance by generating additional fault-like samples, while the classifier is designed to capture local features, long-range temporal dependencies, and diagnostically informative temporal regions. (3) Results: A multidimensional evaluation based on temporal similarity, statistical consistency, and global distribution discrepancy indicates that the generated samples preserve important characteristics of the original signals under the current evaluation protocol. On the augmented LRE dataset, the proposed classifier achieved strong diagnostic performance. In addition, supplementary experiments on the public HIT aero-engine dataset further support the effectiveness of the classifier architecture, its component-wise contribution, and its behavior under imbalanced few-shot settings, while also demonstrating the value of uncertainty-aware prediction. (4) Conclusions: The results provide encouraging evidence that the proposed framework can improve LRE fault diagnosis under data-scarce conditions. However, the present findings should be interpreted within the scope of the available data and evaluation setting. More comprehensive generator-side ablation, broader external validation, and physics-oriented assessment of the generated signals are still needed before stronger conclusions can be made. Full article

Systematically characterizing the time series deformation evolution of open-pit mine slopes is key to revealing their potential instability development and supporting subsequent deformation-level classification. Interferometric Synthetic Aperture Radar (InSAR), by enabling measurement of ground deformation at a global scale approximately every ten days, may hold the key to those interactions. However, atmospheric propagation delays still have a significant impact on deformation calculations, and open-pit mine slopes monitored by InSAR often suffer from low coherence. This noise can obscure nonlinear and transient precursory signatures in deformation time series, reducing the identifiability of key temporal patterns required for automated interpretation. Here, we present a Coherence-conditioned Encoder–Decoder Long Short-Term Memory (CED-LSTM) denoising network for deformation time series. We generate a physics-aware synthetic dataset by modeling coherence-dependent measurement noise and temporally correlated atmospheric delays. The network jointly models deformation time series and coherence, using residual learning and adaptive gated composite loss to preserve deformation trends. It is designed to autonomously extract ground deformation signals from noise in InSAR time series without prior knowledge of where deformation occurs or how it evolves. On the synthetic validation set, the network achieved a root mean square error (RMSE) of 2.2 mm across the validation sequences. Applied to three InSAR datasets over an open-pit mine from March 2019 to March 2022, denoising suppresses noise and stabilizes deformation boundaries, enabling extraction of trend and transient indicators and a data-driven deformation-level score. Using quantile-based thresholds, these scores are then used to produce multi-year deformation-level classification maps. Full article

Video action recognition faces persistent challenges in balancing accuracy with computational efficiency. While state space models, such as Mamba, have emerged with linear complexity advantages, they exhibit inefficiency in capturing critical spatiotemporal dependencies within video data. To address this core limitation, this paper proposes ETMamba, an enhanced architecture built upon the Mamba baseline. The ETMamba achieve performance breakthroughs via three core innovation modules: (1) the Spatiotemporal Feature Preservation module retains complete original spatiotemporal correlations before data flattening, solving the problem of spatiotemporal feature loss; (2) the Efficient Bidirectional Sharing strategy accurately models bidirectional temporal dependencies, enhancing key temporal dynamic information; and (3) the Spatiotemporal Collaborative Modulation mechanism combines global temporal and local spatial information to achieve collaborative capture of long-short term dependencies and fine-grained features. We conduct experiments on multiple benchmark datasets, achieving recognition accuracies of 88.3%, 74.6%, 75.7%, and 98.1% on Kinetics-400, Something-Something V2, HMDB-51, and Breakfast datasets, respectively, while maintaining low to medium computational complexity. Full article

Indonesia is among the most landslide-prone countries in the world, with thousands of fatalities and widespread infrastructure damage recorded over recent decades. Despite this high hazard level, regional-scale landslide monitoring remains constrained by the limitations of conventional bitemporal satellite imagery, which is susceptible to cloud contamination, dependent on precise acquisition timing, and unable to capture the full temporal dynamics of landslide occurrence and recovery. While the LandTrendr (Landsat-based Detection of Trends in Disturbance and Recovery) algorithm has been widely applied for detecting vegetation disturbances such as forest loss and land-use change, its potential for landslide detection in tropical environments has not been sufficiently explored. This study aims to evaluate the applicability of LandTrendr applied to long-term Landsat time series imagery for automated regional-scale landslide detection and mapping in Indonesia. The method integrates temporal segmentation of the Normalized Difference Vegetation Index (NDVI) derived from Landsat imagery spanning 2000–2022 with slope information from the Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) to identify the characteristic drop-recovery spectral signature associated with landslide events. The algorithm was applied and evaluated in two geologically distinct study areas: Lombok, West Nusa Tenggara, and Pasaman, West Sumatra. Detection accuracies of 25.9% by location and 20.3% by area were achieved in Lombok and 76.3% by location and 85.3% by area in Pasaman. The lower accuracy in Lombok is primarily attributed to the predominance of small landslides below the sensor’s spatial resolution and rapid vegetation recovery. The proposed approach demonstrates the unique capability of LandTrendr to model the entire life cycle of a mass movement event, from pre-event stability through abrupt disturbance to ecological recovery within a single unified framework, providing a scalable and cost-effective tool for long-term landslide monitoring applicable to other tropical, landslide-prone regions. Full article

Advances in semiconductor technologies, particularly in power transistors and switching diodes, have enabled higher switching frequencies and converter efficiency, renewing interest in Switched Reluctance Motors (SRMs) for electric vehicles. This work presents a data-driven approach utilizing a Long Short-Term Memory (LSTM) network capable of effectively managing temporal dependencies for estimating rotor position without sensors in SRMs. The motor investigated was custom-designed, subsequently manufactured as a prototype. The LSTM was trained and validated with experimental data collected at various speeds and load conditions. The outcomes demonstrate the model’s strong performance, with a mean squared error (MSE) of ${1.77}^{{°}^2}$, a mean absolute error (MAE) of 1.09°, and 97.35% accuracy. Compared to typical estimation methods such as back-electromotive force (EMF)-based techniques, fuzzy logic, model predictive control, feed-forward neural networks (FFNNs), and back-propagation neural networks (BPNNs), the LSTM stands out as one of the most effective and widely used models. Previous neural networks (NN)-based studies typically report ±5° accuracy, whereas LSTM keeps the error about 1° in this study. This strategy eliminates position sensors, reduces cost and complexity, and enables reliable real-time SRM control. Results indicate that the method has significant potential for electric motor drives, particularly for SRMs. Full article

Viewport prediction is a key component in tile-based 360° video streaming. Existing viewport prediction models based on Long Short-term Memory Networks (LSTM) or Transformer typically output a single deterministic future trajectory through deterministic mapping, which fails to capture the inherent randomness in viewing behavior. Moreover, when encoding trajectory features, such models often map trajectory coordinates directly into a high-dimensional space while neglecting the spatial information inherent in the coordinates themselves. Additionally, they exhibit limitations in capturing cross-modal relationships between visual and trajectory features. To address these issues, this paper proposes DiffVP, a diffusion model for viewport prediction in 360° videos. Under the constraints of viewing historical trajectories and video saliency maps, DiffVP leverages Denoising Diffusion Implicit Models (DDIMs) to model future viewing trajectories in the form of probability distributions, generating diverse and reasonable prediction results. In the denoising network, DiffVP employs Explicit Coordinate-Time Encoding (ECTE) to model the temporal dependencies of trajectories and the spatial relationships among coordinates; moreover, a Coordinate-Aware Saliency Features Fusion (CASF) module is proposed to achieve cross-modal alignment and interactive fusion of saliency and trajectory features. Experimental results on three public datasets demonstrate that DiffVP achieves the best accuracy for 2–5 s viewport prediction without sacrificing the performance of short-term (<1 s) prediction. Full article

Reflectometry-based sensing systems are widely used in industrial monitoring to assess the condition of distributed assets such as cables and transmission lines. In practical sensing environments, however, correlation-based interpretation can become unreliable because of bilinear interference, dispersive propagation, and excitation mismatch, often producing artifact-related responses that lead to unnecessary inspections and reduced decision reliability. This paper proposes a temporal-consistency-based reliability enhancement framework for correlation-driven time–frequency domain reflectometry (TFDR). Instead of replacing the conventional reflectometry pipeline, the proposed method introduces a reliability-estimation layer that evaluates the trustworthiness of correlation responses and suppresses temporally inconsistent artifacts. Multiple complementary descriptors extracted from the reflected signal are jointly analyzed to determine whether a correlation response is propagation-consistent or more likely to arise from non-physical artifacts. Temporal consistency is modeled using a bidirectional long short-term memory (BiLSTM) architecture that captures long-range dependencies along the propagation sequence. Experimental results obtained from cable reflectometry measurements under varying impedance conditions show that the proposed framework effectively suppresses artifact-related correlation responses while preserving physically meaningful reflections required for fault localization. Additional cross-excitation evaluation provides preliminary evidence that the learned temporal-consistency criterion is not tightly coupled to a single excitation waveform. Because the proposed framework operates as a post-processing reliability layer, it can be integrated into existing reflectometry-based monitoring systems without the modification of the sensing hardware or excitation scheme. Full article

Electricity demand in the United States is steadily increasing due to rapid technological growth, especially the expansion of AI data centers and electric vehicles, which are becoming major power consumers. At the same time, rising renewable energy integration, changing weather patterns, and the deployment of battery energy storage systems are increasing variability and complexity in grid operations. These evolving conditions require advanced forecasting methods to ensure reliability and efficiency, as traditional statistical and machine learning models struggle with nonlinear and temporal dependencies. To address these challenges, this study proposes a hybrid deep learning framework that combines convolutional neural networks, long short-term memory, and bidirectional LSTM models to forecast electricity generation across both conventional and renewable energy sources. The framework incorporates seasonal-trend decomposition using loess to extract trend, seasonal, and residual components, enhancing the learning of multi-scale temporal patterns. A key contribution of this work is the development of a unified, source-specific forecasting system in which each energy source is assigned its best-performing hybrid architecture. The proposed framework achieves superior accuracy, with the CNN-Bi-LSTM model yielding the best total power results (MAPE 2.60%, RMSE 13,745 MWh, MAE 9542 MWh), while Bi-LSTM models excel for wind, biomass, geothermal, and nuclear. This enables scalable, high-precision national-level forecasting. Full article

Accurate building load forecasting is essential for smart grid and energy management, yet nonlinearity, non-stationarity, and multi-scale characteristics of load data challenge traditional methods. To address these issues, we propose a hybrid deep learning framework, CEEMDAN-MultiScale-CNN-BiLSTM-MultiAttention. First, Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) decomposes the load sequence into intrinsic mode functions (IMFs), mitigating mode mixing and complexity. Then, a MultiScale Convolutional Neural Network extracts multi-scale local features from each IMF. A Bidirectional Long Short-Term Memory network captures bidirectional temporal dependencies, and a Multi-Attention mechanism dynamically emphasizes critical time steps and feature channels, enhancing interpretability and prediction. The framework is validated on the Building Data Genome Project 2 dataset, achieving a Mean Absolute Percentage Error (MAPE) of 2.6464% and a coefficient of determination R² of 0.8999, outperforming mainstream methods across multiple metrics. The main contributions are: (1) a hybrid framework integrating CEEMDAN, multi-scale feature extraction, and attention mechanisms to handle nonlinearity and non-stationarity; (2) a MultiScale-CNN to capture multi-scale temporal features and adapt to multi-frequency components; (3) a Multi-Attention mechanism to dynamically focus on key time steps and channels, improving accuracy and robustness. This work provides an effective solution for building load forecasting in complex energy systems. Full article