Search Results (1,613)

Vegetation dynamics are complexly influenced by multiple factors such as climate, human activities, and topography. In recent years, the frequency, intensity, and diversity of human activities have increased, placing substantial pressure on the growth of vegetation. Arid and semi-arid regions are particularly sensitive to climate change, and climate change and large-scale ecological restoration have led to significant changes in the dynamic of dryland vegetation. However, few studies have explored the nonlinear relationships between these factors and vegetation dynamic. In this study, we integrated trend analysis (using the Mann–Kendall test and Theil–Sen estimation) and machine learning algorithms (XGBoost-SHAP model) based on long time-series remote sensing data from 2001 to 2020 to quantify the nonlinear response patterns and threshold effects of bioclimatic variables, topographic features, soil attributes, and anthropogenic factors on vegetation dynamic. The results revealed the following key findings: (1) The kNDVI in the study area showed an overall significant increasing trend (p < 0.01) during the observation period, of which 26.7% of the area showed a significant increase. (2) The water content index (Bio 23, 19.6%), the change in land use (15.2%), multi-year average precipitation (pre, 15.0%), population density (13.2%), and rainfall seasonality (Bio 15, 10.9%) were the key factors driving the dynamic change of vegetation, with the combined contribution of natural factors amounting to 64.3%. (3) Among the topographic factors, altitude had a more significant effect on vegetation dynamics, with higher altitude regions less likely to experience vegetation greening. Both natural and anthropogenic factors exhibited nonlinear responses and interactive effects, contributing to the observed dynamic trends. This study provides valuable insights into the driving mechanisms behind the condition of vegetation in arid and semi-arid regions of China and, by extension, in other arid regions globally. Full article

Amidst the escalating demand for high-precision structural health monitoring in large-scale engineering applications, carbon fiber-reinforced polymer fiber Bragg grating (CFRP-FBG) sensors have emerged as a pivotal technology for fatigue life evaluation, owing to their exceptional sensitivity and intrinsic immunity to electromagnetic interference. A time-series predictive architecture based on long short-term memory (LSTM) networks is developed in this work to facilitate intelligent fatigue life assessment of structures subjected to complex cyclic loading by capturing and modeling critical spectral characteristics of CFRP-FBG sensors, specifically the side-mode suppression ratio and main-lobe peak-to-valley ratio. To enhance model robustness and generalization, Principal Component Analysis (PCA) was employed to isolate the most salient spectral features, followed by data preprocessing via normalization and model optimization through the integration of the Adam optimizer and Dropout regularization strategy. Relative to conventional Backpropagation (BP) neural networks, the LSTM model demonstrated a substantial improvement in predicting the side-mode suppression ratio, achieving a 61.62% reduction in mean squared error (MSE) and a 34.99% decrease in root mean squared error (RMSE), thereby markedly enhancing robustness to outliers and ensuring greater overall prediction stability. In predicting the peak-to-valley ratio, the model attained a notable 24.9% decrease in mean absolute error (MAE) and a 21.2% reduction in root mean squared error (RMSE), thereby substantially curtailing localized inaccuracies. The forecasted confidence intervals were correspondingly narrower and exhibited diminished fluctuation, highlighting the LSTM architecture’s enhanced proficiency in capturing nonlinear dynamics and modeling temporal dependencies. The proposed method manifests considerable practical engineering relevance and delivers resilient intelligent assistance for the seamless implementation of CFRP-FBG sensor technology in structural health monitoring and fatigue life prognostics. Full article

Stock price forecasting remains a persistent challenge in time series analysis due to complex inter-stock relationships and dynamic textual signals such as financial news. While Graph Neural Networks (GNNs) can model relational structures, they often struggle with capturing higher-order dependencies and are sensitive to noise. Moreover, sentiment signals are typically aggregated using fixed time windows, which may introduce temporal bias. To address these issues, we propose DASF-Net (Diffusion-Aware Sentiment Fusion Network), a multimodal framework that integrates structural and textual information for robust prediction. DASF-Net leverages diffusion processes over two complementary financial graphs—one based on industry relationships, the other on fundamental indicators—to learn richer stock representations. Simultaneously, sentiment embeddings extracted from financial news using FinBERT are aggregated over an empirically optimized window to preserve temporal relevance. These modalities are fused via a multi-head attention mechanism and passed to a temporal forecasting module. DASF-Net integrates daily stock prices and news sentiment, using a 3-day sentiment aggregation window, to forecast stock prices over daily horizons (1–3 days). Experiments on 12 large-cap S&P 500 stocks over four years demonstrate that DASF-Net outperforms competitive baselines, achieving up to 91.6% relative reduction in Mean Squared Error (MSE). Results highlight the effectiveness of combining graph diffusion and sentiment-aware features for improved financial forecasting. Full article

Rapid urbanization has intensified global settlement density, significantly increasing the importance of urban street environmental quality, which profoundly affects residents’ physical and psychological well-being. Traditional methods for evaluating urban environmental quality have largely overlooked dynamic perceptual changes occurring throughout the day, resulting in incomplete assessments. To bridge this methodological gap, this study presents an innovative approach combining advanced deep learning techniques with time-series street view imagery (SVI) analysis to systematically quantify spatio-temporal variations in the perceived environmental quality of pedestrian-oriented streets. It further addresses two central questions: how perceived environmental quality varies spatially across sections of a pedestrian-oriented street and how these perceptions fluctuate temporally throughout the day. Utilizing Golden Street, a representative living street in Shanghai’s Changning District, as the empirical setting, street view images were manually collected at 96 sampling points across multiple time intervals within a single day. The collected images underwent semantic segmentation using the DeepLabv3+ model, and emotional scores were quantified through the validated MIT Place Pulse 2.0 dataset across six subjective indicators: “Safe,” “Lively,” “Wealthy,” “Beautiful,” “Depressing,” and “Boring.” Spatial and temporal patterns of these indicators were subsequently analyzed to elucidate their relationships with environmental attributes. This study demonstrates the effectiveness of integrating deep learning models with time-series SVI for assessing urban environmental perceptions, providing robust empirical insights for urban planners and policymakers. The results emphasize the necessity of context-sensitive, temporally adaptive urban design strategies to enhance urban livability and psychological well-being, ultimately contributing to more vibrant, secure, and sustainable pedestrian-oriented urban environments. Full article

Selecting an appropriate model for discharge simulation remains a fundamental challenge in modeling. While artificial neural networks (ANNs) have been widely accepted due to detecting streamflow patterns, they require large datasets for efficient training. However, when short-term datasets are available, training ANNs becomes problematic. Autoregressive integrated moving average (ARIMA) models offer a promising alternative; however, severe volatility, nonlinearity, and trends in hydrological time series can still lead to significant errors. To address these challenges, this study introduces a new adaptive hybrid model, ARIMA-iGARCH, designed to account volatility, variance inconsistency, and nonlinear behavior in short-term hydrological datasets. We apply the model to four hourly discharge time series from the Schwarzbach River at the Nauheim gauge in Hesse, Germany, under the assumption of normally distributed residuals. The results demonstrate that the specialized parameter estimation method achieves lower complexity and higher accuracy. For the four events analyzed, R² values reached 0.99, 0.96, 0.99, and 0.98; RMSE values were 0.031, 0.091, 0.023, and 0.052. By delivering accurate short-term discharge predictions, the ARIMA-iGARCH model provides a basis for enhancing water resource planning and flood risk management. Overall, the model significantly improves modeling long memory, nonlinear, nonstationary shifts in short-term hydrological datasets by effectively capturing fluctuations in variance. Full article

Accurate short-term weather forecasting plays a vital role in disaster response, agriculture, and energy management, where timely and reliable predictions are essential for decision-making. Graph neural networks (GNNs), known for their ability to model complex spatial structures and relational data, have achieved remarkable success in meteorological forecasting by effectively capturing spatial dependencies among distributed weather stations. However, most existing GNN-based approaches rely on pairwise station connections, limiting their capacity to represent higher-order spatial interactions. Moreover, their dependence on supervised learning makes them vulnerable to spatial heterogeneity and temporal non-stationarity. This paper introduces a novel spatial–temporal pretraining framework, Hypergraph-enhanced Meteorological Pretraining (HyMePre), which combines hypergraph neural networks with self-supervised learning to model high-order spatial dependencies and improve generalization across diverse climate regimes. HyMePre employs a two-stage masking strategy, applying spatial and temporal masking separately, to learn disentangled representations from unlabeled meteorological time series. During forecasting, dynamic hypergraphs group stations based on meteorological similarity, explicitly capturing high-order dependencies. Extensive experiments on large-scale reanalysis datasets show that HyMePre outperforms conventional GNN models in predicting temperature, humidity, and wind speed. The integration of pretraining and hypergraph modeling enhances robustness to noisy data and improves generalization to unseen climate patterns, offering a scalable and effective solution for operational weather forecasting. Full article

Federated learning (FL) enables multiple participants to collaboratively train models while efficiently mitigating the issue of data silos. However, large-scale heterogeneous data distributions result in inconsistent client objectives and catastrophic forgetting, leading to model bias and slow convergence. To address the challenges under non-independent and identically distributed (non-IID) data, we propose DPAO-PFL, a Dynamic Parameter-Aware Optimization framework that leverages continual learning principles to improve Personalized Federated Learning under non-IID conditions. We decomposed the parameters into two components: local personalized parameters tailored to client characteristics, and global shared parameters that capture the accumulated marginal effects of parameter updates over historical rounds. Specifically, we leverage the Fisher information matrix to estimate parameter importance online, integrate the path sensitivity scores within a time-series sliding window to construct a dynamic regularization term, and adaptively adjust the constraint strength to mitigate the conflict overall tasks. We evaluate the effectiveness of DPAO-PFL through extensive experiments on several benchmarks under IID and non-IID data distributions. Comprehensive experimental results indicate that DPAO-PFL outperforms baselines with improvements from $5.41\%$ to $30.42\%$ in average classification accuracy. By decoupling model parameters and incorporating an adaptive regularization mechanism, DPAO-PFL effectively balances generalization and personalization. Furthermore, DPAO-PFL exhibits superior performance in convergence and collaborative optimization compared to state-of-the-art FL methods. Full article

Ships play a crucial role in modern society, serving purposes such as marine transportation, tourism, and exploration. Malfunctions or defects in the main engine, which is a core component of ship operations, can disrupt normal functionality and result in substantial financial losses. Consequently, early fault diagnosis of abnormal engine conditions is critical for effective maintenance. In this paper, we propose a deep hybrid model for fault diagnosis of ship main engines, utilizing exhaust gas temperature data. The proposed model utilizes both time-domain features (TDFs) and time-series raw data. In order to effectively extract features from each type of data, two distinct feature extraction networks and an attention module-based classifier are designed. The model performance is evaluated using real-world cylinder exhaust gas temperature data collected from the large ship low-speed two-stroke main engine. The experimental results demonstrate that the proposed method outperforms conventional methods in fault diagnosis accuracy. The experimental results demonstrate that the proposed method improves fault diagnosis accuracy by 6.146% compared to the best conventional method. Furthermore, the proposed method maintains superior performanceeven in noisy environments under realistic industrial conditions. This study demonstrates the potential of using exhaust gas temperature using a single sensor signal for data-driven fault detection and provides a scalable foundation for future multi-sensor diagnostic systems. Full article

Accurate ship berthing behavior prediction (BBP) is essential for enabling collision warnings and support decision-making. Existing methods based on Automatic Identification System (AIS) data perform well in the task of ship trajectory prediction over long time-series and large scales, but struggle with addressing the fine-grained and highly dynamic changes in berthing scenarios. Therefore, the accuracy of BBP remains a crucial challenge. In this paper, a novel BBP method based on Light Detection and Ranging (LiDAR) data is proposed. To test its feasibility, a comprehensive dataset is established by conducting on-site collection of berthing data at Dalian Port (China) using a shore-based LiDAR system. This dataset comprises equal-interval data from 77 berthing activities involving three large ships. In order to find a straightforward architecture to provide good performance on our dataset, a cascading network model combining convolutional neural network (CNN), a bi-directional gated recurrent unit (BiGRU) and bi-directional long short-term memory (BiLSTM) are developed to serve as the baseline. Experimental results demonstrate that the baseline outperformed other commonly used prediction models and their combinations in terms of prediction accuracy. In summary, our research findings help overcome the limitations of AIS data in berthing scenarios and provide a foundation for predicting complete berthing status, therefore offering practical insights for safer, more efficient, and automated management in smart port systems. Full article

Amid the growing debate over how cryptocurrencies are reshaping global finance, this study explores the nexus between Bitcoin, Brent Crude Oil, Gold and the U.S. Dollar Index. We used a time-varying vector autoregressive (tvVAR) model to examine the connection among these four assets during the Trump (2017–2020) and Biden (2021–2024) governments. The 48-week return forecast of the Bitcoin–Gold correlation was also conducted by using the Bayesian Structural Time Series (BSTS) model. Results indicate that Bitcoin was the most volatile asset, while the U.S. Dollar remained the least volatile under both regimes. Under Trump, U.S. Dollar significantly influenced Oil and Bitcoin while Bitcoin and Gold were negatively linked to Oil and positively associated with U.S. Dollar. An inverse relationship between Bitcoin and Gold also emerged. Under Biden, Bitcoin, Gold, and U.S. Dollar all significantly affected Oil with Bitcoin showing a positive impact. Bitcoin and Gold remained negatively correlated though not significantly, and the Dollar maintained positive ties with both. Forecasts show a positive link between Bitcoin and Gold in the coming year. However, Bitcoin does not exhibit consistent characteristics of a safe-haven asset during the U.S. presidential transitions examined, largely due to its high volatility and unstable correlations with a traditional safe-haven asset, Gold. This study contributes to the understanding of shifting relationships between digital and traditional assets across political regimes. Full article

During the in-hospital optimization of dopaminergic dosage for Parkinson’s disease, drug-induced dyskinesias emerge as a common side effect. Wrist-worn devices present a substantial opportunity for continuous movement recording and the supportive identification of these dyskinesias. To bridge the gap between dyskinesia assessment and machine learning-enabled detection, the recorded information requires meaningful data representations. This study evaluates and compares two distinct representations of sensor data: a task-dependent, semantically grounded approach and automatically extracted large-scale time-series features. Each representation was assessed on public datasets to identify the best-performing machine learning model and subsequently applied to our own collected dataset to assess generalizability. Data representations incorporating semantic knowledge demonstrated comparable or superior performance to reported works, with peak F₁ scores of 0.68. Generalization to our own dataset from clinical practice resulted in an observed F₁ score of 0.53 using both setups. These results highlight the potential of semantic movement data analysis for dyskinesia detection. Dimensionality reduction in accelerometer-based movement data positively impacts performance, and models trained with semantically obtained features avoid overfitting. Expanding cohorts with standardized neurological assessments labeled by medical experts is essential for further improvements. Full article

Accurate prediction of drone motion within structured urban air corridors is essential for ensuring safe and efficient operations in Urban Air Mobility (UAM) systems. Although real-world Remote Identification (Remote ID) regulations require drones to broadcast critical flight information such as velocity, access to large-scale, high-quality broadcast data remains limited. To address this, this study leverages a Digital Twin (DT) framework to augment Remote ID spatio-temporal broadcasts, emulating the sensing environment of dense urban airspace. Using Remote ID data, we propose BiDGCNLLM, a hybrid prediction framework that integrates a Bidirectional Graph Convolutional Network (BiGCN) with Dynamic Edge Weighting and a reprogrammed Large Language Model (LLM, Qwen2.5–0.5B) to capture spatial dependencies and temporal patterns in drone speed trajectories. The model forecasts near-future speed variations in surrounding drones, supporting proactive conflict avoidance in constrained air corridors. Results from the AirSUMO co-simulation platform and a DT replica of the Cranfield University campus show that BiDGCNLLM outperforms state-of-the-art time series models in short-term velocity prediction. Compared to Transformer-LSTM, BiDGCNLLM marginally improves the $R^2$ by 11.59%. This study introduces the integration of LLMs into dynamic graph-based drone prediction. It shows the potential of Remote ID broadcasts to enable scalable, real-time airspace safety solutions in UAM. Full article

This research presents a novel method for detecting building roof types using deep learning models based on height profiles from high-resolution digital surface models. While deep learning has proven effective in digit, handwritten, and time series classification, this study focuses on the emerging and crucial area of height profile detection for building roof type classification. We propose an innovative approach to automatically generate, classify, and detect building roof types using height profiles derived from normalized digital surface models. We present three distinct methods to detect seven roof types from two height profiles of the building cross-section. The first two methods detect the building roof type from two-dimensional (2D) height profiles: two binary images and a two-band spectral image. The third method, vector-based, detects the building roof type from two one-dimensional (1D) height profiles represented as two 1D vectors. We trained various one- and two-dimensional convolutional neural networks on these 1D and 2D height profiles. The DenseNet201 network could directly detect the roof type of a building from two height profiles stored as a two-band spectral image with an average accuracy of 97%, even in the presence of consecutive chimneys, dormers, and noise. The strengths of this approach include the generation of a large, detailed, and storage-efficient labeled height profile dataset, the development of a robust classification method using both 1D and 2D height profiles, and an automated workflow that enhances building roof type detection. Full article

In response to the challenge of multi-objective optimal scheduling and efficient solution of hydropower stations under large-scale renewable energy integration, this study develops a multi-objective optimization model with the dual goals of maximizing total power generation and minimizing the variance of residual load. Four complementarity evaluation indicators are used to analyze the wind–solar complementarity characteristics. Building upon this foundation, Hyper-dominance Evolutionary Algorithm (HEA)—capable of efficiently solving high-dimensional problems—is introduced for the first time in the context of wind–solar–hydropower integrated scheduling. The case study results show that the HEA performs better than the benchmark algorithms, with the best mean Hypervolume and Inverted Generational Distance Plus across nine Walking Fish Group (WFG) series test functions. For the hydro-wind-solar scheduling problem, HEA obtains Pareto frontier solutions with both maximum power generation and minimal residual load variance, thus effectively solving the multi-objective scheduling problem of the hydropower system. This work provides a valuable reference for modeling and efficiently solving the multi-objective scheduling problem of hydropower in the context of emerging power systems. This work provides a valuable reference for the modeling and efficient solution of hydropower multi-objective scheduling problems in the context of emerging power systems. Full article

The top oil temperature of a transformer is a vital sign reflecting its operational condition. The accurate prediction of this parameter is essential for evaluating insulation performance and extending equipment lifespan. At present, the prediction of oil temperature is mainly based on single-feature prediction. However, it overlooks the influence of other features. This has a negative effect on the prediction accuracy. Furthermore, the training dataset is often made up of data from a single transformer. This leads to the poor generalization of the prediction. To tackle these challenges, this paper leverages large-scale data analysis and processing techniques, and presents a transformer top oil temperature prediction model that combines multiple models. The Convolutional Neural Network was applied in this method to extract spatial features from multiple input variables. Subsequently, a Long Short-Term Memory network was employed to capture dynamic patterns in the time series. Meanwhile, a Transformer encoder enhanced feature interaction and global perception. The spatial characteristics extracted by the CNN and the temporal characteristics extracted by LSTM were further integrated to create a more comprehensive representation. The established model was optimized using the Whale Optimization Algorithm to improve prediction accuracy. The results of the experiment indicate that the maximum RMSE and MAPE of this method on the summer and winter datasets were 0.5884 and 0.79%, respectively, demonstrating superior prediction accuracy. Compared with other models, the proposed model improved prediction performance by 13.74%, 36.66%, and 43.36%, respectively, indicating high generalization capability and accuracy. This provides theoretical support for condition monitoring and fault warning of power equipment. Full article