Search Results (440)

Accurate forecasting of multivariate time series is essential for energy management, grid optimisation, and policy planning. This study presents a hybrid deep learning and Transformer-based forecasting framework for predicting hourly electricity consumption across Turkey using nationwide data from Energy Exchange Istanbul (EPİAŞ) between 2018 and 2025. The dataset comprises 15 variables representing diverse energy sources and market indicators, including consumption, generation, and the market-clearing price (MCP). The proposed hybrid model integrates Long Short-Term Memory (LSTM), Bidirectional LSTM (BLSTM), and Gated Recurrent Unit (GRU) layers to capture both short- and long-term temporal dependencies, while a Transformer model leveraging multi-head self-attention mechanisms is used for comparison. All models were trained using standardised preprocessing, a 24 h lookback window, and optimised hyperparameters via GridSearchCV. Experimental results reveal that the hybrid model achieved the best overall performance, with MAE = 464.01, RMSE = 663.39, and R² = 0.9902, significantly outperforming the baseline and Transformer models. The Transformer demonstrated robust long-horizon learning capability (R² = 0.9257) but at a higher computational cost. These results confirm that combining multiple recurrent architectures enhances predictive accuracy and stability for large-scale, real-time energy forecasting. The proposed framework offers a reliable foundation for smart grid operations, demand prediction, and data-driven energy policy development. Full article

The frequency of extreme weather events has become higher, and electricity consumption has also become more complex. These changes increase the risk of overload in distribution transformers (DTs), and this risk threatens the stability and reliability of the power grid. Existing methods have significant limitations. Traditional static threshold methods (based on DGA gas ratios and electrical signal thresholds) fail to consider temporal changes and complex links between factors, while modern machine learning models lack cause–effect relationships over time and clear ways to describe uncertainty. With such motivations, this paper proposes a causal-enhanced hybrid framework, which combines Long Short-Term Memory (LSTM) networks and Random Forest (RF) algorithms. The framework uses causal Seasonal Trend decomposition using Loess (STL) to reveal load patterns at different time scales. The mutual information index and spatiotemporal graph convolutional network (ST-GCN) are used to explore nonlinear relations and reveal how temperature affects load changes. The LSTM model captures time dependence in load series, and the Bayesian optimized Random Forest is used to solve the problem of data imbalance and quantify uncertainty. In addition, the framework constructs an early warning system that combines data from many sources in real time. Test results show that the proposed algorithm exhibits excellent performance in multi-source data environments. Full article

To address the challenges of insufficient transformer winding fault samples and the effective fusion of heterogeneous multi-source data, this study proposes an intelligent fault diagnosis method based on a time–frequency diffusion model and ConvNeXt-1D. First, data augmentation is performed on the original signals using the time–frequency diffusion model. Through a forward noise injection and reverse denoising process, the limited time-series samples are expanded. By alternately applying time-domain noise addition and frequency-domain blurring, the signals are jointly enhanced in the time–frequency domain, improving sample diversity and feature representation. Next, a ConvNeXt-1D network is constructed for multi-scale feature extraction and fault classification, incorporating an attention mechanism to efficiently fuse multi-source features and achieve precise fault identification. Finally, the proposed method is validated using dynamic model experiments. The results indicate that under typical fault conditions—such as inter-turn short circuits, winding deformation, and arc discharge—the proposed method achieves a diagnostic accuracy of 99.23 ± 0.29%. Compared with other classical models, the proposed approach demonstrates stronger classification capability and higher stability under small-sample data conditions. Full article

Smart city decision-making increasingly relies on heterogeneous urban data sources. Dense traffic sensor streams provide continuous quantitative measurements, while citizen-generated textual reports offer event-driven contextual information. However, integrating these modalities remains challenging due to temporal misalignment, textual sparsity, and semantic noise. This paper investigates multi-modal learning for traffic congestion severity prediction through an experimental integration of open traffic sensor data (METR-LA: Los Angeles, USA) and citizen-generated textual reports (NYC 311: New York City, USA). Congestion severity is formulated as a four-class classification task derived from traffic speed measurements. We propose an end-to-end framework that combines: (i) sensor time-series encoding using a GRU-based temporal encoder, (ii) textual representation learning using a BERT-based encoder, (iii) a symmetric time-window alignment strategy (±Δ) to associate irregular reports with sensor time steps, and (iv) multiple fusion architectures, including early fusion, late fusion, and a cross-attention module for cross-modal interaction modeling. Experiments on publicly available datasets show that multi-modal early fusion achieves the best overall performance (Accuracy = 0.8283, Macro-F1 = 0.8231) compared to uni-modal baselines. In the studied cross-city setting with sparse and weakly aligned textual signals, the proposed cross-attention fusion does not outperform the strong sensor-only baseline, suggesting that the sensor modality dominates when cross-modal signal strength is limited. These results highlight both the potential and the practical constraints of multi-modal fusion in heterogeneous smart-city environments, emphasizing the importance of alignment design, modality relevance, and transparent experimental validation. Full article

Time-series interferometric synthetic aperture radar (TS-InSAR) has become a widely used technique for monitoring surface deformation with high spatial and temporal resolution. The recent rise in cloud-based InSAR platforms has significantly accelerated the production of interferograms. However, the accuracy of deformation inversion remains limited by fundamental issues affecting interferogram quality, including temporal and spatial decorrelation and phase unwrapping errors. These degrading effects are most pronounced in vegetated, desert, and snow-covered terrains, which are common in active tectonic zones and thereby exert a major impact on the quality of the unwrapped phase. Traditional quality control methods are inefficient or inadequate for large-scale analysis, and discarding low-quality data reduces the inversion accuracy. To address these limitations, we developed a deep learning-based approach to automatically assess interferogram quality and integrate it into the time-series InSAR inversion workflow. We utilized Sentinel-1 interferograms generated by the COMET-LiCSAR system as the primary data source. Based on this dataset, we developed a multi-stage selection strategy for interferogram quality control, integrating loop phase closure analysis, statistical indicators (including coherence and phase standard deviation), and manual verification. As a result, we constructed a high-quality labeled dataset comprising approximately 20,000 samples. An improved ConvNeXt-InSAR model was designed and trained to automatically quantify the quality of each pixel in individual interferograms. The model generates pixel-wise quality maps, which are then incorporated as weight constraints in the time-series InSAR network inversion. The proposed method was applied to the interseismic deformation reconstruction in the central-southern Tibetan Plateau region. This study highlights the potential of deep learning-based interferogram quality assessment in facilitating large-scale, automated time-series InSAR processing. Full article

Real-time processing demands for massive IoT sensor data necessitate reliance on distributed microservice systems within edge clusters. However, pinpointing the root cause of anomalies within these edge microservice clusters poses a critical challenge for intelligent IoT operation and maintenance. To address the issue, a spatio-temporal graph propagation model ST-GraphRCA is proposed for root cause analysis in IoT edge environments. Our approach begins by resolving the fundamental issue of time-series asynchrony across distributed multi-source metrics. A PCA-DTW hybrid feature extraction method is introduced with a dynamic alignment strategy to mitigate the effects of random network delays and data deformation without requiring prior synchronization. Subsequently, ST-GraphRCA constructs a stream-based forward propagation graph based on the flow conservation principle. By integrating dynamic edge weights with node-level input–output anomaly scores, ST-GraphRCA precisely infers fault propagation pathways and identifies potential root cause candidates through causal reasoning. Finally, a topology-constrained high-utility mining algorithm filters these candidates. Using a constraint matrix, the algorithm filters out unreachable service combinations to locate low-frequency and high-risk root causes. Experimental results indicate that ST-GraphRCA achieves an F1-Score of 0.89, outperforming existing methods. In resource-constrained edge scenarios, its average localization time is merely 238.8 ms, representing a six-fold improvement over key benchmarks. Thus, ST-GraphRCA not only provides an efficient anomaly fault tracing solution for large-scale IoT systems but also offers technical support for the intelligent operation and maintenance of distributed microservice systems. Full article

Enhancing cereal grain quality while maintaining yield stability represents a pressing global challenge for sustainable agricultural development. Optimizing grain quality in cereal crops, which account for more than 60% of global dietary energy, relies heavily on managing nitrogen dynamics during the heading and grain-filling stages. Late-stage nitrogen application (from heading to early grain-filling stages) optimizes the temporal dynamics of nitrogen supply and exhibits substantial regulatory potential in mediating the yield–quality trade-off. Nitrogen availability can profoundly influence source–sink dynamics, carbon–nitrogen metabolic coordination, and the biosynthesis of storage reserves. This systematic review consolidates current understanding of the molecular and physiological mechanisms by which late-stage nitrogen application affects grain development and final quality in cereals, with a particular focus on major cereal crops including wheat, rice, and malting barley, which represent contrasting quality objectives and nitrogen management requirements. At the physiological level, late-stage nitrogen application delays functional leaf senescence, sustains photosynthetic carbon assimilation capacity, facilitates assimilate transport and partition to developing grains, and optimizes the accumulation dynamics and compositional profiles of starch and protein. At the molecular level, this review elucidates the sequential regulatory cascades governing nitrogen signal perception and transduction, the coordinated transcriptional networks underlying carbon–nitrogen metabolic crosstalk, and the expression dynamics of genes encoding starch biosynthetic enzymes and storage proteins. Integrating those recent research advances, this review also highlights several critical challenges currently facing the field. To address these challenges, we delineate promising avenues for future research including constructing time-series multi-omics frameworks, employing genome-editing technologies to functionally validate key regulatory genes and integrating artificial intelligence and big data analytics. The goal of this review is to establish a theoretical basis for precision nitrogen management strategies designed to optimize cereal crop production, targeting high yield, superior quality, and improved nitrogen use efficiency concurrently. Full article

The accurate calculation of excavation volume is critical for open-pit mine planning and management. Traditional methods are often inefficient and constrained by operational conditions. In contrast, digital surface model (DSM) differential analysis using stereophotogrammetry enables rapid acquisition of excavation volume, which holds significant value for retrospective excavation process. However, the actual mining process is not a simple matter of “excavation” or “backfilling”, but rather a complex mining pattern involving repeated excavation as new coal seams are exposed. This study utilized multi-source stereo remote sensing data (ZY-3, GF-7 satellite and UAV data) to construct a high-precision DSM time series spanning 2013 to 2025, focusing on analyzing the topographical evolution patterns of three representative mining pits. Research indicates that constructing DSMs during summer and autumn yields higher conformity with actual terrain, RMSE = 1.67 m and ME = −0.07 m. To address diverse mining patterns, we propose two calculation methods: the Cumulative Method (CM), which captures iterative excavation-backfilling cycles, and the First-Last Subtraction Method (FLSM), which mitigates cumulative DSM errors during continuous excavation. For phased mining operations, a hybrid method combining both approaches yields optimal results. Validation in three typical pits showed relative calculation errors of 1.36%, −0.49%, and 1.68%, respectively. The study indicates that the surface morphology changes in open-pit mines exhibit distinct non-linear characteristics. The method proposed herein not only enhances computational accuracy but also provides technical support for tracing historical coal excavation volumes. Full article

Accurate monitoring of aboveground biomass (AGB) is essential for forest carbon accounting and climate change mitigation, yet signal saturation and the treatment of forest landscapes as biophysically homogeneous entities remain significant barriers to high-fidelity mapping. This study implements an ecologically integrated model that leverages forest-type specific (coniferous vs. broadleaf) to enhance regional AGB retrieval. By refining established data fusion techniques with structural and compositional parameters, this approach seeks to mitigate systematic biases often found in generic regional assessments. Compared with 360 geo-referenced subplots, our stratified Support Vector Regression (SVR) model significantly outperformed non-classified counterparts, achieving an R² of 0.76 and a reduced RMSE of 18.48 Mg/ha. This refined precision enabled a nuanced time-series analysis (2013–2020), revealing that while regional AGB increased from 157.13 to 192.23 Mg/ha, this trajectory was punctuated by a distinct sub-regional growth plateau between 2016 and 2018. By correlating these trends with disturbance data, we identified a 11.27% biomass decline in southwestern sectors linked to a tripling of burned area, pinpointing intensified fire regimes as the primary driver overriding recovery-driven carbon gains. These findings demonstrate that harmonizing multi-sensor signals with functional forest differentiation provides the necessary sensitivity to track carbon resilience, offering a scalable and robust tool for operational forest management and global carbon cycle research. Full article

Accurate and timely mapping of paddy rice is essential for agricultural management, food security, and climate-resilient policy. However, high-precision mapping remains challenging in subtropical monsoon regions due to persistent cloud cover, long revisit intervals, and striping noise, which compromise satellite data quality and availability. To address these limitations, a rice mapping framework suitable for different geographical environments was developed based on a random forest (RF) by combining time-series harmonic analysis (HANTS) with Sentinel-1 and Sentinel-2 multi-source data. To address these limitations, a rice mapping classification algorithm for different geographical environments was developed by combining Harmonic Analysis of Time Series (HANTS) with Sentinel-1/2 multi-source data. The research obtained annual maps of single-season and double-season rice in the research area from 2019 to 2024, with a spatial resolution of 10 m. The results indicated that the Sentinel-1, Sentinel-2, GEE, and HANTS algorithm can effectively support the yearly mapping of single- and double-season paddy rice in Anhui Province, China. The resultant paddy rice map has a high accuracy with overall accuracies exceeding 92% and Kappa coefficients above 0.84. HANTS effectively captures key phenological features of paddy rice, and it can especially enhance the discrimination between single- and double-season rice; compared to existing rice mapping products, the proposed approach reduces classification errors by an average of 3.92% in six major rice-producing cities, each with cultivation areas exceeding 1 million hectares; spatial correlation analysis indicates substantial heterogeneity in rice cultivation patterns across northern, central, and southern Anhui, associated with both biophysical and anthropogenic factors. These results indicate that integrating phenological data with machine learning can enhance the accuracy of long-term, high-resolution crop monitoring, and annual rice maps will offer valuable support for food security assessment, water resource management, and policy planning. Full article

Accurate estimation of solar radiation is of great significance for solar energy development and climate research. However, in China, the scarcity and uneven distribution of observation stations often cause deep learning models to overfit and suffer from accuracy degradation under small-sample conditions. To address this issue, this paper proposes a deep learning framework that integrates transfer learning and multi-scale time series modeling for predicting hourly global solar radiation at target meteorological sites. The method employs representation learning and clustering to select source domain sites with similar climatic characteristics. It integrates wavelet transform convolution, depthwise separable convolution, and a Transformer encoder–decoder to achieve multi-scale feature extraction and long-term dependency modeling. Experimental results demonstrate that the model achieved a coefficient of determination (R²) of 0.9710 in tests conducted in the Ningxia region. It maintained good predictive performance even in a cold-start scenario with only one month of training data and exhibited stable accuracy across all four seasons, effectively mitigating seasonal bias. This provides a reliable solution for solar radiation estimation in data-scarce regions, and its modeling approach can also be extended to other climate-related time series prediction tasks. Full article

Forecasting post-seismic landslide displacement is challenged by the difficulty in distinguishing short-term acceleration from creep and the risk of spatiotemporal leakage. To address this, an interpretable deep-learning framework is developed, integrating SBAS-InSAR time series with an Attention-enhanced Gated Recurrent Unit (Attention-GRU). Prior to modeling, a multi-stage preprocessing strategy, including empirical mode decomposition, is applied to mitigate noise and delineate active deformation zones. Unlike standard architectures, the model’s temporal attention mechanism adaptively amplifies critical precursory acceleration phases. Furthermore, a strict landslide-object-based partitioning strategy is employed to rigorously mitigate spatiotemporal leakage. The framework was evaluated in the Le’an Town landslide cluster using multi-source data. Targeting identified hazardous regions, the method achieved an R² of 0.93 and reduced MAPE by 42.7% relative to the SVR baseline. This reflects a location-specific predictive capability, within active zones rather than regional generalization. SHapley Additive exPlanations (SHAP) further confirmed the model captures physical relationships, such as sensitivity to 25–35° slopes and vegetation degradation. Ultimately, the proposed framework offers a transparent, physically interpretable tool for operational hazard mitigation. Full article

Land subsidence poses a persistent challenge to Tianjin, a major coastal city in China, with implications for urban infrastructure and sustainable development. This study examines the spatiotemporal evolution of ground subsidence in Tianjin from 2003 to 2024 using multi-source SAR observations from Envisat ASAR (C-band), ALOS PALSAR (L-band), and Sentinel-1 (C-band). Surface deformation was derived using SBAS-InSAR with atmospheric phase correction. Due to limitations in data availability, SAR observations are temporally discontinuous; therefore, the long-term subsidence evolution was reconstructed by integrating multi-sensor deformation rates through a model-based time-series fitting approach. The results show pronounced subsidence during 2003–2010 in inland districts such as Wuqing, Beichen, Jinnan, and Jinghai, with maximum rates exceeding 50 mm/yr. After 2017, regional subsidence rates generally declined, while localized deformation became increasingly concentrated in coastal reclamation areas of the Binhai New Area, particularly around Dongjiang Port and Fuzhuang. Spatial and temporal patterns of subsidence exhibit clear correspondence with changes in groundwater use intensity and phases of urban construction and land reclamation. These observations suggest a transition in dominant subsidence controls over time. The results provide a long-term observational perspective on subsidence evolution in Tianjin and offer a geospatial basis for land-use planning and infrastructure risk assessment in coastal cities. Full article

For assessing forest resource quality and carbon sequestration, both optical and synthetic aperture radar (SAR) remote sensing data have been widely used to map forest aboveground carbon storage (AGC), demonstrating considerable potential across diverse forest types. However, the fusion approaches between SAR and optical data remain technically challenging, particularly when combining multi-source optical and multi-temporal SAR datasets. In this study, multiple optical datasets with varying spatial resolutions and spectral bands (Landsat-9, Sentinel-2, GF-6 PMS, and GF-6 WFV) and time-series Sentinel-1 data acquired within the same year were employed to develop an optical–SAR fusion framework for mapping forest AGC in mixed broadleaf–coniferous forests. Firstly, a multi-level collaborative fusion strategy (MLC) was developed using multi-source optical data by integrating the strengths of both pixel-level and feature-level fusion. Subsequently, a multi-temporal SAR combining approach was designed based on seasonal variation patterns using one-year time-series Sentinel-1 data. Finally, an optical–SAR modeling approach was established to map forest AGC using multiple machine learning models combined with the sequential forward feature selection method. The results demonstrate that the proposed MLC fused method for multi-source optical data offers significant advantages in enhancing estimation accuracy and improving model robustness. Furthermore, when multi-temporal Sentinel-1 data were integrated with the MLC-fused optical data, the optical–SAR collaborative approach further improved the coefficient of determination (R²), effectively mitigating the saturation effect commonly observed in optical data. The highest performance was achieved using spring-acquired multi-temporal Sentinel-1 data within the SVR model, yielding an R² of 0.69 and reducing rRMSE to 18.03%. It is indicated that an appropriate fusing strategy for integrating optical and SAR data can substantially enhance both accuracy and reliability in mapping forest AGC. Full article

Water quality monitoring is critical for public health, ecology, and economic sustainability, but traditional methods are limited by temporal-spatial coverage and cost, failing to meet real-time assessment needs. Deep learning for water quality prediction is often hindered by high complexity and noise in raw time series. This study aims to address the high complexity and noise of hydrological time series by proposing a prediction framework integrating sliding window feature enhancement, principal component analysis (PCA), and a two-layer regularized gated recurrent unit (TLR-GRU). The core goal is to achieve high-precision real-time prediction of four key water quality parameters (dissolved oxygen (DO), ammonia nitrogen (${\mathrm{NH}}_3\text{-}\mathrm{N}$), total phosphorus (TP), and total nitrogen (TN)) for aquaculture and irrigation. Sample entropy (SampEn, $m=2$, $r=0.2 \times \mathrm{std}\left(X\right)$), a univariate complexity metric capturing intra-series pattern repetition, quantifies time series regularity, showing sliding windows reduce SampEn by filtering transient noise while retaining ecological patterns. This optimization synergizes with TLR-GRU’s regularization (L2, Dropout) to avoid overfitting. A total of 4970 water quality records (2020–2023, 4 h sampling interval) were collected from a monitoring station in a typical aquaculture-irrigated water body. After dimensionality reduction via PCA, experimental results demonstrate that the TLR-GRU model outperforms six state-of-the-art deep learning models (e.g., TLD-LSTM, WaveNet) on both the base dataset and the sliding window-enhanced dataset. On the latter, DO and TP test set $R^2$ rise from 0.82 to 0.93 and 0.81 to 0.92, with RMSE decreasing by 49.4% and 55.6%, respectively. This framework supports water resource management, applicable to rivers and lakes beyond aquaculture. Future work will optimize the model and integrate multi-source data. Full article