Search Results (113)

Efficient wind power forecasting is critical in achieving large-scale integration of wind energy in modern electricity systems. On the other hand, limited availability of wealthy, long-term historical data of wind power generation for many sites of interest often challenges the training of tailored forecasting models, which, in turn, introduces uncertainty concerning the anticipated operational status of similar early-life, or even prospective, wind farm projects. To that end, this study puts forward a spatiotemporal, national-level forecasting exercise as a means of addressing wind power data scarcity in Greece. It does so by developing a hybrid wavelet-enhanced deep learning model that leverages long-term historical data from a reference site located in central Greece. The model is optimized for 24-h day-ahead forecasting, using a hybrid architecture that incorporates discrete wavelet transform for feature extraction, with deep neural networks for spatiotemporal learning. Accordingly, the model’s generalization is evaluated across a number of geographically distributed sites of different quality wind potential, each constrained to only one year of available data. The analysis compares forecasting performance between the original and target sites to assess spatiotemporal robustness of the model without site-specific retraining. Our results demonstrate that the developed model maintains competitive accuracy across data-scarce locations for the first 12 h of the day-ahead forecasting horizon, designating, at the same time, distinct performance patterns, dependent on the geographical and wind potential quality dimensions of the examined areas. Overall, this work underscores the feasibility of leveraging data-rich regions to inform forecasting in under-instrumented areas and contributes to the broader discourse on spatial generalization in renewable energy modeling and planning. Full article

Objectives: To address the challenges of subjectivity, misdiagnosis and underdiagnosis in post-traumatic stress disorder (PTSD), this study proposes an objective auxiliary diagnostic method based on P300 signals. Existing studies largely rely on conventional P300 features, lacking the systematic integration of event-related potential (ERP) priors and facing limitations in spatiotemporal feature modeling. Methods: Using common spatiotemporal pattern (CSTP) analysis and quantitative evaluation, we revealed significant spatiotemporal differences in P300 signals between PTSD patients and healthy controls. ERP prior information was then extracted and integrated into a hybrid architecture combining transformer encoders and a convolutional neural network (CNN), enabling joint modeling of long-range temporal dependencies and local spatial patterns. Results: The proposed P300 spatiotemporal transformer-CNN (P300-STTCNet) achieved a classification accuracy of 93.37% in distinguishing PTSD from healthy controls, markedly outperforming traditional approaches. Conclusions: Significant spatiotemporal differences in P300 signals exist between PTSD and healthy control groups. The P300-STTCNet model effectively captures PTSD-related spatiotemporal features, demonstrating strong potential for electroencephalogram-based objective auxiliary diagnosis. Full article

The global pursuit of sustainable development amplifies renewable energy’s strategic importance, positioning wind power as a vital modern grid component. Accurate wind forecasting is essential to counter inherent volatility, enabling robust grid operations, security protocols, and optimization strategies. Such predictive precision directly governs wind energy systems’ stability and sustainability. This research introduces a novel spatio-temporal hybrid model integrating convolutional neural networks (CNN), bidirectional long short-term memory (BiLSTM), and graph convolutional networks (GCN) to extract temporal patterns and meteorological dynamics (wind speed, direction, temperature) across 134 wind turbines. Building upon conventional methods, our architecture captures turbine spatio-temporal correlations while assimilating multivariate meteorological characteristics. Addressing data integrity compromises from equipment failures and extreme weather-which undermine data-driven models-we implement Wasserstein GAN (WGAN) for generative missing-value interpolation. Validation across severe data loss scenarios (30–90% missing values) demonstrates the model’s enhanced predictive capacity. Rigorous benchmarking confirms significant accuracy improvements and reduced forecasting errors. Full article

The background of the article refers to the diagnosis of focal liver lesions (FLLs) through contrast-enhanced ultrasound (CEUS) based on the integration of spatial and temporal information. Traditional computer-aided diagnosis (CAD) systems predominantly rely on static images, which limits the characterization of lesion dynamics. This study aims to assess the effectiveness of Transformer-based architectures in enhancing CAD performance within the realm of liver pathology. The methodology involved a systematic comparison of deep learning models for the analysis of CEUS images and videos. For image-based classification, a Hybrid Transformer Neural Network (HTNN) was employed. It combines Vision Transformer (ViT) modules with lightweight convolutional features. For video-based tasks, we evaluated a custom spatio-temporal Convolutional Neural Network (CNN), a CNN with Long Short-Term Memory (LSTM), and a Video Vision Transformer (ViViT). The experimental results show that the HTNN achieved an outstanding accuracy of 97.77% in classifying various types of FLLs, although it required manual selection of the region of interest (ROI). The video-based models produced accuracies of 83%, 88%, and 88%, respectively, without the need for ROI selection. In conclusion, the findings indicate that Transformer-based models exhibit high accuracy in CEUS-based liver diagnosis. This study highlights the potential of attention mechanisms to identify subtle inter-class differences, thereby reducing the reliance on manual intervention. Full article

This study proposes a novel method to address the instability issues in multi-target bearing-only tracking for passive sonar systems. Utilizing a particle filter-guided on-site training mechanism, the complex multi-classification task is simplified into binary classification (target vs. non-target) by assigning an independent tracker to each target. This enables simultaneous on-site training and deployment of the neural network for tracking. A hybrid CNN-BiLSTM network is constructed: the Convolutional Neural Network (CNN) enhances target feature extraction and non-target discrimination, while the Bidirectional Long Short-Term Memory (BiLSTM) models spatiotemporal dependencies. Their synergy improves trajectory continuity and smoothness. Under simulated conditions, the proposed method reduces the minimum required SNR for stable tracking to −31.78 dB, a significant improvement over the −29.69 dB required by pure particle filtering methods. The average tracking error is also reduced from 0.61° to 0.34°. Both simulations and sea trial data demonstrate that the method maintains stable tracking even during target trajectory crossings, significantly enhancing multi-target tracking accuracy in complex underwater acoustic environments. Full article

Karst springs play a critical strategic role in regional economic and ecological sustainability, yet their spatiotemporal heterogeneity and hydrological complexity pose substantial challenges for flow prediction. This study proposes FMD-mGTO-BiGRU-KAN, a four-stage hybrid deep learning architecture for daily spring flow prediction that integrates multi-feature signal decomposition, meta-heuristic optimization, and interpretable neural network design: constructing an Feature Mode Decomposition (FMD) decomposition layer to mitigate modal aliasing in meteorological signals; employing the improved Gorilla Troops Optimizer (mGTO) optimization algorithm to enable autonomous hyperparameter evolution, overcoming the limitations of traditional grid search; designing a Bidirectional Gated Recurrent Unit (BiGRU) network to capture long-term historical dependencies in spring flow sequences through bidirectional recurrent mechanisms; introducing Kolmogorov–Arnold Networks (KAN) to replace the fully connected layer, and improving the model interpretability through differentiable symbolic operations; Additionally, residual modules and dropout blocks are incorporated to enhance generalization capability, reduce overfitting risks. By integrating multiple deep learning algorithms, this hybrid model leverages their respective strengths to adeptly accommodate intricate meteorological conditions, thereby enhancing its capacity to discern the underlying patterns within complex and dynamic input features. Comparative results against benchmark models (LSTM, GRU, and Transformer) show that the proposed framework achieves 82.47% and 50.15% reductions in MSE and RMSE, respectively, with the NSE increasing by 8.01% to 0.9862. The prediction errors are more tightly distributed, and the proposed model surpasses the benchmark model in overall performance, validating its superiority. The model’s exceptional prediction ability offers a novel high-precision solution for spring flow prediction in complex hydrological systems. Full article

Against the backdrop of the “dual circulation” development pattern and the in-depth advancement of the Regional Comprehensive Economic Partnership (RCEP), the interconnection between China and global financial markets has significantly intensified. The spatio-temporal correlation risks faced in cross-border investment activities have become highly complex, posing a severe challenge to traditional investment risk prediction methods. Existing research has three limitations: first, traditional analytical tools struggle to capture the dynamic spatio-temporal correlations among financial markets; second, mainstream deep learning models lack the ability to directly output interpretable economic parameters; third, the uncertainty of model prediction results has not been systematically quantified for a long time, leading to a lack of credibility assessment in practical applications. To address these issues, this study constructs a spatio-temporal graph convolutional neural network panel regression model (STGCN-PDR) that incorporates uncertainty quantification. This model innovatively designs a hybrid architecture of “one layer of spatial graph convolution + two layers of temporal convolution”, modeling the spatial dependencies among global stock markets through graph networks and capturing the dynamic evolution patterns of market fluctuations with temporal convolutional networks. It particularly embeds an interpretable regression layer, enabling the model to directly output regression coefficients with economic significance, significantly enhancing the decision-making reference value of risk prediction. By designing multi-round random initialization perturbation experiments and introducing the coefficient of variation index to quantify the stability of model parameters, it achieves a systematic assessment of prediction uncertainty. Empirical results based on stock index data from 20 countries show that compared with the benchmark models, STGCN-PDR demonstrates significant advantages in both spatio-temporal feature extraction efficiency and risk prediction accuracy, providing a more interpretable and reliable quantitative analysis tool for cross-border investment decisions in complex market environments. Full article

The rapid growth in the number of motor vehicles has exacerbated traffic congestion. The occurrence of congestion not only poses significant challenges for traffic management authorities but also severely impacts residents’ travel and daily routines. Against this backdrop, predicting traffic flow can provide crucial insights for anticipating changing traffic patterns. Therefore, this paper proposes a novel hybrid deep learning architecture (CNN-LSTM-GRU) for highway traffic flow prediction that integrates spatiotemporal and meteorological dimensions. Our approach constructs a multidimensional feature matrix encompassing temporal sequences, spatial correlations, and weather conditions. Convolutional Neural Networks (CNN) are employed to capture spatial patterns, while Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks jointly model temporal dependencies. Through systematic hyperparameter tuning and step-length optimization, we validate the model using real-world traffic data from a provincial highway network. The experimental evaluation analyzes the following two critical dimensions: (1) holiday vs. non-holiday traffic patterns, and (2) the impact of weather data integration. Comparative analysis reveals that our hybrid model demonstrates superior prediction accuracy over standalone LSTM, GRU, and their CNN-based counterparts (CNN-LSTM, CNN-GRU). Full article

To overcome the limitations of traditional convolutional and recurrent neural networks in capturing spatio-temporal dynamics in flow fields on unstructured grids, this study proposes a novel Spatio-Temporal Graph Neural Network (ST-GNN) model that integrates a Graph Neural Network (GNN) with a Long Short-Term Memory (LSTM) network. The GNN component captures spatial dependencies among irregular grid nodes via message passing, while the LSTM component models temporal evolution through gated memory mechanisms. This hybrid framework enables the joint learning of spatial and temporal features in complex flow systems. Two variants of ST-GNN, namely, GCN-LSTM and GAT-LSTM, were developed and evaluated using the SUBOFF AFF-8 benchmark dataset. The results show that GAT-LSTM achieved higher accuracy than GCN-LSTM, with average relative errors of 2.51% for velocity and 1.43% for pressure at the 1000th time step. Both models achieved substantial speedups over traditional CFD solvers, with GCN-LSTM and GAT-LSTM accelerating predictions by approximately 350 and 150 times, respectively. These findings position ST-GNN as an efficient and accurate alternative for spatio-temporal flow modeling on unstructured grids, advancing data-driven fluid dynamics. Full article

Desertification poses critical threats to agricultural productivity and socio-economic stability, particularly in vulnerable regions like Thatta and Badin districts of Sindh, Pakistan. Traditional monitoring methods lack the accuracy and temporal resolution needed for effective early warning systems. This study presents a novel Spatio-Temporal Desertification Predictor (STDP) framework that integrates deep learning with next-generation satellite imaging for time-series desertification forecasting. The proposed encoder–decoder architecture combines Convolutional Neural Networks (CNNs) for spatial feature extraction from high-resolution satellite imagery with modified Long Short-Term Memory (LSTM) networks enhanced by multi-head attention to capture temporal dependencies. Environmental variables are fused through an adaptive data integration layer, and hyperparameter optimization is employed to enhance model performance for edge computing deployment. Experimental validation on a 15-year satellite dataset (2010–2024) demonstrates superior performance with MSE = 0.018, MAE = 0.079, and $R^2=0.94$, outperforming traditional CNN-only, LSTM-only, and hybrid baselines by 15–20% in prediction accuracy. The framework forecasts desertification trends through 2030, providing actionable signals for environmental management and policy-making. This work advances the integration of AI with satellite-based Earth observation, offering a scalable path for real-time environmental monitoring in IoT and edge computing infrastructures. Full article

Driven by the increasing global population and rapid urbanization, aircraft noise pollution has emerged as a significant environmental challenge, impeding the sustainable development of the aviation industry. Traditional noise prediction methods are limited by incomplete datasets, insufficient spatiotemporal consistency, and poor adaptability to complex meteorological conditions, making it difficult to achieve precise noise management. To address these limitations, this study proposes a novel noise prediction framework based on a hybrid Convolutional Neural Network–Bidirectional Long Short-Term Memory–Attention (CNN–BiLSTM–Attention) model. By integrating multi-source data, including meteorological parameters (e.g., temperature, humidity, wind speed) and aircraft trajectory data (e.g., altitude, longitude, latitude), the framework achieves high-precision prediction of aircraft noise. The Haversine formula and inverse distance weighting (IDW) interpolation are employed to effectively supplement missing data, while spatiotemporal alignment techniques ensure data consistency. The CNN–BiLSTM–Attention model leverages the spatial feature extraction capabilities of CNNs, the bidirectional temporal sequence processing capabilities of BiLSTMs, and the context-enhancing properties of the attention mechanism to capture the spatiotemporal characteristics of noise. The experimental results indicate that the model’s predicted mean value of 68.66 closely approximates the actual value of 68.16, with a minimal difference of 0.5 and a mean absolute error of 0.89%. Notably, the error remained below 2% in 91.4% of the prediction rounds. Furthermore, ablation studies revealed that the complete CNN–BiLSTM–AM model significantly outperformed single-structure models. The incorporation of the attention mechanism was found to markedly enhance both the accuracy and generalization capability of the model. These findings highlight the model’s robust performance and reliability in predicting aviation noise. This study provides a scientific basis for effective aviation noise management and offers an innovative solution for addressing noise prediction problems under data-scarce conditions. Full article

Water quality deterioration poses a critical threat to ecological security and sustainable development, particularly in rapidly urbanizing regions. To enable proactive environmental management, this study develops a novel hybrid deep learning model, the NGO-CNN-GRU, for high-precision time-series water quality prediction in the Xijiang River Basin, China. The model integrates a Convolutional Neural Network (CNN) for spatial feature extraction and a Gated Recurrent Unit (GRU) for temporal dependency modeling, with hyperparameters optimized via the Northern Goshawk Optimization (NGO) algorithm. Using historical water quality (pH, DO, CODMn, NH3-N, TP, TN) and meteorological data (precipitation, temperature, humidity) from 11 monitoring stations, the model achieved exceptional performance: test set R² > 0.986, MAE < 0.015, and RMSE < 0.018 for total nitrogen prediction (Xiaodong Station case study). Across all stations and indicators, it consistently outperformed baseline models (GRU, CNN-GRU), with average R² improvements of 12.3% and RMSE reductions up to 90% for NH3-N predictions. Spatiotemporal analysis further revealed significant pollution gradients correlated with anthropogenic activities in the Pearl River Delta. This work provides a robust tool for real-time water quality early warning systems and supports evidence-based river basin management. Full article

Wireless sensor networks (WSNs) are widely used in environmental monitoring, industrial automation, and smart cities. The Hybrid Energy-Efficient Distributed (HEED) protocol is a popular clustering algorithm designed to prolong network lifetime by balancing energy consumption among sensor nodes. However, HEED relies on simple heuristics for cluster-head (CH) selection, which may not fully exploit the complex spatiotemporal patterns in node energy and topology. This paper introduces a novel protocol, Deep Learning–Hybrid Energy-Efficient Distributed (DL-HEED), which, for the first time, integrates a Graph Neural Network (GNN) into the clustering process. By leveraging the relational structure of WSNs and a comprehensive set of node and network features—including residual energy, node degree, spatial position, and signal strength—DL-HEED enables intelligent, context-aware, and adaptive CH selection. DL-HEED leverages the relational structure of WSNs through deep learning, enabling more adaptive and energy-efficient cluster head selection than traditional heuristic-based protocols. Extensive simulations demonstrate that DL-HEED significantly outperforms classic HEED achieving up to 60% improvement in the network lifetime and energy efficiency as the network size increases. This work establishes DL-HEED as a robust, scalable, and practical solution for next-generation WSN deployments, marking a substantial advancement in the application of deep learning to resource-constrained IoT environments. Full article

Currently, the short-term accurate prediction of multi-region taxi pick-up demand often adopts methods that integrate graph neural networks with temporal modeling. However, most models focus solely on node features during the learning process, neglecting or simplifying edge features. This study adopts a hybrid prediction framework, Edge-GATv2-LSTM, which integrates an edge-aware attention-based graph neural network (Edge-GATv2) with a temporal modeling component (LSTM). The framework not only models spatial interactions among regions via GATv2 and temporal evolution via LSTM but also incorporates edge features into the attention computation structure, jointly representing them with node features. This enables the model to perceive both node attributes and the strength of inter-regional relationships during attention weight calculation. Experiments are conducted based on real-world taxi order data from Ningbo City, and the results demonstrate that the adopted Edge-GATv2-LSTM model exhibits favorable performance in terms of pick-up demand prediction accuracy. Specifically, the model achieves the lowest RMSE and MAE of 3.85 and 2.86, respectively, outperforming all baseline methods and confirming its effectiveness in capturing spatiotemporal demand patterns. This research can provide decision-making support for taxi drivers, platform operators, and traffic management departments—for example, by offering a reference basis for optimizing taxi pick-up route planning when vehicles are unoccupied. Full article

Soil moisture is a key factor in the exchange of energy and matter between the soil and atmosphere, playing a vital role in the hydrological cycle and agricultural management. Traditional monitoring methods are limited in achieving large-scale, real-time observations, while deep learning offers new avenues to model the complex nonlinear relationships between spectral features and soil moisture content. This study focuses on the Wei-Ku Oasis in Xinjiang, using multi-source remote sensing data (Landsat series and Sentinel-1) and in situ multi-layer soil moisture measurements. The BOSS feature selection algorithm was applied to construct 46 feature parameters, including vegetation indices, soil indices, and microwave indices, and to identify optimal variable sets for each depth. Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and their hybrid model (CNN-LSTM) were used to build soil moisture inversion models at various depths. Their performances were systematically compared on both training and testing sets, and the optimal model was used for spatiotemporal mapping. The results show that the CNN-LSTM-based multi-depth soil moisture inversion model achieved superior performance, with the 0–10 cm model showing the highest accuracy and a testing R² of 0.64, outperforming individual models. The testing R² values for the soil moisture inversion models at depths of 10–20 cm, 20–40 cm, and 40–60 cm were 0.59, 0.54, and 0.59, respectively. According to the mapping results, soil moisture in the 0–60 cm profile of the Wei-Ku Oasis exhibited a vertical gradient, increasing with depth. Spatially, soil moisture was higher in the central oasis and lower toward the periphery, forming a “center-high, edge-low” pattern. This study provides a high-accuracy method for multi-layer soil moisture remote sensing in arid regions, offering valuable data support for oasis water resource management and precision irrigation planning. Full article