Search Results (2,435)

Against the backdrop of the global energy transition toward deep electrification, the natural gas industry faces challenges, including increased load forecasting uncertainty and frequent extreme weather impacts. To enhance natural gas load forecasting accuracy and support system resilience planning, this study constructs a forecasting model based on quadratic decomposition and hybrid deep learning, incorporating an electricity substitution coefficient to characterize the coupling substitution effect between electricity and natural gas. Under the basic scenario, the VMD-WPD-TCN-BiGRU model is proposed. It employs variational mode decomposition and wavelet packet denoising for secondary signal denoising, combined with a time-series convolutional network and bidirectional gated recurrent unit to extract temporal features. Experiments demonstrate that, compared to mainstream methods such as CNN, BiLSTM, SVM, and XGBoost, this model achieves statistically significant reductions in MSE (11.11–96.21%), MAE (0.89–76.50%), and MAPE (4.10–67.94%), significantly improving forecasting accuracy. In the deep electrification scenario, the introduction of the electricity substitution coefficient further optimizes peak load forecasting for system pressure days under extreme low temperatures, elevating the overall R² to 0.9905 in the deep electrification scenario. Research indicates that the proposed model not only effectively improves the accuracy of short-term natural gas load forecasting but also provides quantitative support for enterprises to plan peak-shaving facilities, optimize pipeline networks, and respond to extreme weather emergencies in data silo environments. This contributes to strengthening the adaptability and long-term resilience of natural gas systems during the energy transition, thereby supporting the sustainable development of energy infrastructure. Full article

Forecasting financial time series (FTS) is essential for analyzing and understanding the dynamics of financial markets. Traditional recurrent neural network (RNN) models often suffer from low prediction accuracy on non-stationary and abruptly changing data, as their gating mechanisms struggle to capture evolving trends in FTS. This paper introduces variational mode decomposition (VMD) and multifractal analysis to enhance the gating mechanism of the gated recurrent unit (GRU). By quantifying the changing characteristics of FTS, the proposed model dynamically adjusts the gating weights. In addition, a state fusion strategy is employed to improve the utilization efficiency of historical information. Experiments are conducted using daily data of the SSE 50, CSI 300, and CSI 1000 indices, spanning from 4 January 2002, to 26 December 2025. The results demonstrate that, compared to traditional models, the proposed model better captures the evolving characteristics of FTS and achieves higher prediction accuracy. Full article

Grammatical error correction (GEC) for Chinese learner English is still dominated by sentence-level modeling, which limits discourse-level consistency and weakens adaptation to learner-specific error profiles. From an instructional perspective, these limitations also reduce the value of automated feedback as a basis for spaced-repetition instructional structures sensitive to individual differences. This study proposes a symmetry-aware dual-encoder architecture for context-aware GEC in Chinese learner English. A context encoder captures preceding-sentence information, while a source encoder integrates BERT-based semantic representations with Bi-GRU-based syntactic features for the current sentence. A gated decoder performs asymmetric fusion of local and contextual evidence. To better reflect corpus-level tendencies in Chinese learner English, a CLEC-informed augmentation strategy generates synthetic errors using empirical category frequencies as a coarse sampling prior. Experiments on CoNLL-2014, JFLEG, and CLEC show consistent improvements over strong neural baselines in F_0.5 and GLEU under the current desktop-oriented implementation setting. Nevertheless, the integration of BERT, dual encoders, and gated decoding introduces non-negligible computational overhead, and the present system is therefore better suited to desktop writing-support scenarios than to strict real-time or large-scale online deployment. The proposed framework thus provides a practical technical basis for personalized grammar feedback and for future spaced-repetition instructional designs in ESL writing support. Full article

Remaining Useful Life (RUL) prediction is a key component of prognostics and health management for modern industrial systems. While deep learning methods have significantly improved prediction accuracy, many existing approaches rely on large neural networks that are difficult to deploy on resource-constrained edge devices. This study presents a deployment-oriented evaluation of compact neural networks for RUL prediction using the NASA C-MAPSS turbofan engine benchmark. Two lightweight hybrid architectures, CNN–GRU and CNN–TCN, were developed with approximately 28k–32k parameters to represent realistic models for CPU-based edge inference. A systematic experimental analysis was conducted across all four C-MAPSS subsets (FD001–FD004), which represent increasing levels of operational and fault complexity. In addition to baseline performance, two post-training compression techniques (i.e., global unstructured magnitude pruning and dynamic INT8 quantization) were evaluated. To assess real deployment behavior, inference latency was measured on both a high-performance Intel x86 workstation and a resource-constrained ARM platform. Results show that CNN–GRU generally achieves higher predictive accuracy, whereas CNN–TCN provides more consistent and lower inference latency due to its convolution-only temporal modeling. Unstructured pruning can yield modest improvements in prediction accuracy, suggesting a regularization effect, but it does not reliably reduce model size or latency on standard CPUs due to the overhead associated with pruning masks. Dynamic quantization substantially reduces model size (particularly for CNN–GRU) while preserving predictive accuracy; however, it increases runtime latency because of additional quantization and dequantization operations. These findings demonstrate that compression techniques commonly used for large models do not necessarily translate into deployment benefits for already compact RUL architectures and highlight the importance of hardware-aware evaluation when designing edge prognostics systems. Full article

Short-term forecasting of solar radiation is essential for the efficient operation of solar energy systems. This study presents a neural network-based approach for short-term solar radiation forecasting using a hybrid framework that integrates temporal characteristics with weather-based features. The proposed model combines a Gated Recurrent Unit (GRU) to capture short-term temporal dynamics, a Transformer Encoder, and a Multilayer Perceptron (MLP) to integrate these representations for final prediction. Key meteorological variables, including temperature, humidity, and wind speed, are incorporated along with engineered time-related features such as lagged values, rolling statistics, and cyclical time-of-day encodings. The results demonstrate that the hybrid model effectively integrates sequential learning and feature interaction, leading to improved forecasting accuracy. The proposed approach achieves a test Mean Absolute Error (MAE) of 0.056, Root Mean Square Error (RMSE) of 0.086, and coefficient of determination (R²) of 0.92, outperforming benchmark models such as AutoRegressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), GRU, and Extreme Gradient Boosting (XGBoost). The model maintains stable performance across cross-validation folds, multiple forecasting horizons, and varying weather conditions. These findings indicate that the proposed framework provides a reliable and practical solution for accurate short-term solar radiation forecasting, supporting real-time solar energy management and renewable energy system optimization. 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

Classroom humor is an important instructional strategy that enhances teaching effectiveness and improves student engagement. However, its automatic detection remains challenging due to the strong contextual dependency and implicit semantic shifts that characterize humorous expressions in teaching discourse. Conventional pretrained language models capture global semantics but often fail to focus on the subtle humor anchors that trigger incongruity. To address this issue, we propose T-HumorAGSA, a cognitive-inspired classroom teacher humor language detection model. The model employs BERT for contextualized semantic encoding, followed by a Gated Anchor-Guided Self-Attention (AGSA) mechanism that adaptively amplifies anchor-related features responsible for humor generation. A bidirectional gated recurrent unit (BiGRU) layer is further integrated to model long-range temporal dependencies within teaching utterances. T-HumorAGSA is evaluated on five datasets, including SemEval 2021 Task 7-1a, ColBERT, CCL2018, CCL2019 and the self-constructed teacher humor language dataset (T-Humor), demonstrating consistently strong performance. For instance, it achieves 0.9874 F1 on ColBERT and 0.9508 F1 on SemEval 2021 Task 7-1a, both outperforming the best baseline models. On the T-Humor dataset, the model attains a high F1 score of 0.9895, validating its capacity to detect subtle humorous cues in instructional discourse. The results demonstrate that the proposed model delivers excellent performance in classroom humor detection. Full article

In the context of the coordinated development of high-proportion renewable energy integration and alternating current/direct current (AC/DC) hybrid transmission, the sending-end power grid faces challenges such as decreased system strength, contracted stability boundaries, and difficulties in covering high-risk operating conditions. This paper proposes a new renewable energy scale impact analysis method that integrates “typical scenario construction-scale ladder comparison–prediction-driven time series injection” in response to the operational constraints of AC/DC transmission. In terms of method implementation, firstly, a two-layer typical scenario system is constructed under unified transmission constraints and fixed grid boundaries: A regular benchmark scenario covers the main operating range, and a set of high-risk scenarios near the boundaries is obtained through multi-objective intelligent search, which is then refined through clustering to form a computable stress-test scenario library. Here, the boundary scenarios are generated by a multi-objective search that simultaneously drives multiple key section load rates towards their limits, subject to AC power-flow feasibility and operational constraints, and the resulting Pareto candidates are reduced into a compact stress-test library by clustering. Secondly, a ladder scenario with increasing renewable energy scale is constructed, and cross-scale comparisons are carried out within the same scenario system to extract the scale effect and critical laws of key safety indicators. Finally, data resampling and Gated Recurrent Unit multi-step prediction are introduced to generate wind power output time series, enabling the temporal mapping of prediction results to scenario injection quantities, and constructing a closed-loop input interface of “prediction–scenario–grid indicators”. The results demonstrate that the proposed hierarchical framework, under unified AC/DC export constraints, can effectively construct a compact stress-test scenario library with enhanced boundary-risk coverage and can reveal how transient voltage security evolves across renewable expansion scales. By coupling boundary-oriented scenario construction, cross-scale comparable assessment, and forecasting-driven time series injection, the framework improves engineering interpretability and practical applicability compared with conventional scenario sampling/reduction workflows. For the forecasting module, the Gated Recurrent Unit (GRU) model achieves MAPE = 8.58% and RMSE = 104.32 kW on the test set, outperforming Linear Regression (LR)/Random Forest (RF)/Support Vector Regression (SVR) in multi-step ahead prediction. Full article

The nonlinearity and the extreme variability of seismic signals makes the detection of earthquakes difficult. Although the conventional deep-learning models can be used to extract useful features, they cannot be used in early-warning systems due to their non-interpretability. In this study, a Bidirectional Long Short-Memory network with an attention system (Bi-LSTM-Attn) is proposed to detect seismic events using the ConvNetQuake dataset. To improve the quality of data, the entire preprocessing pipeline, such as signal filtering, segmentation, normalization, and noise reduction is employed. The model was optimized using hyperparameter tuning of sequence length, learning rates, and attention weighting to achieve the best number of true-positive detections and a minimum number of false alarms. The accuracy, precision and recall, F1-score, MSE, and ROC curves were used to assess the performance and the attention weight visualization allowed interpreting the model. It is proven through experiments that the Bi-LSTM-Attn model provides more credible and comprehensible forecasting in relation to baseline LSTM and GRU models. Making the high-accuracy classification and the transparent decision behavior, the approach will increase the level of trust to the automated seismic surveillance, as well as help to build the reliable global networks of earthquake early-warnings. 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

Background/Objectives: Type 2 diabetes (T2D) develops gradually over many years through a prolonged preclinical phase, yet traditional static risk scores often fail to capture these dynamic metabolic trajectories. We propose a longitudinal deep learning framework to predict the 10-year risk of Type 2 diabetes onset defined by comprehensive ADA criteria by modeling the physiological acceleration of routine clinical biomarkers. Methods: Utilizing an 18-year longitudinal dataset from the community-based Korean Genome and Epidemiology Study (KoGES) cohort, we selected $N=4354$ participants with complete follow-up records, ensuring high data integrity without requiring synthetic data augmentation. We constructed a 3-dimensional tensor of 21 non-invasive clinical variables spanning a 6-year observation window. To resolve the inherent precision-recall trade-offs of individual models, we developed a stacking ensemble that integrates Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) architectures via a logistic regression meta-learner. To evaluate the added value of longitudinal modeling, we compared this dynamic framework against a static XGBoost baseline that only saw the most recent data. Results: Evaluated on an independent test set ($n=874$), the ensemble significantly outperformed baseline models, achieving an overall accuracy of 0.90 (95% CI: 0.88–0.92) and an AUROC of 0.94 (95% CI: 0.93–0.95). By harmonizing LSTM’s sensitivity and GRU’s precision, the model yielded an exceptional Positive Predictive Value (PPV) of 0.97, a sensitivity of 0.80, and a specificity of 0.98. Conclusions: This framework provides a highly accurate, resource-efficient triage instrument for T2D screening, thereby reducing unnecessary clinical alerts and improving screening efficiency. Full article

Accurate forecasting of mining-induced surface deformation is critical for coal-mine safety assessment and hazard mitigation. InSAR deformation time series are often short, temporally sparse, and strongly nonlinear. These characteristics can make purely data-driven predictors unreliable in small-sample settings. To address this issue, we propose a generation–prediction–interpretation framework that combines generative augmentation with physics-informed forecasting. We first develop a TCN-TimeGAN model to synthesize high-fidelity deformation sequences and expand the training set. Recurrent modules in the generator and discriminator are replaced with causal TCN residual blocks, and a temporal self-attention layer is further stacked on top of the TCN backbone to adaptively reweight informative time steps. We then construct a physics-informed Kolmogorov–Arnold Network, termed PI-KAN. Subsidence-consistency and smoothness priors are embedded in the learning objective to promote physically plausible predictions while retaining spline-based interpretability. Experiments on SBAS-InSAR deformation series from the Guqiao coal mine show that the framework achieves an RMSE of 0.825 mm and an R² of 0.968. It outperforms TGAN-KAN, CNN-BiGRU, and BiGRU under the same evaluation protocol. Visualizations of the learned spline-based edge functions further reveal stronger nonlinear responses for lagged inputs closer to the forecast horizon, providing interpretable evidence of short-term temporal sensitivity under sparse observations. Full article

This study uses multidimensional indicators of macroeconomics, supply and demand, cost, and market microstructure to construct an intelligent framework integrated with optimized Exponentially Weighted Moving Average (EWMA) denoising for price forecasting and black early warning for egg futures in China from 2014 to 2023. Black early warning serves as a non-parametric early warning method that identifies abnormal price increases and falls based on historical fluctuation thresholds. As the first livestock future contract listed in China, accurate egg price forecasting is crucial for risk prevention and market control and regulation. First, LASSO regression was used to screen the core driving factors of egg futures prices. Nine key indicators were identified and input into the hybrid Temporal Convolutional Network–Gated Recurrent Unit (TCN-GRU) prediction model. To address the high-frequency noise in the original price series, two-dimensional optimization was performed on traditional EWMA denoising to achieve more adaptive noise filtering. By applying the black early warning method, the obtained future egg price fluctuations were more consistent with the actual situation. In addition, empirical analysis of multi-horizon forecasting and early warning for t + 1, t + 5, and t + 10 was carried out to further verify the model’s prediction accuracy. The results show that compared with the single TCN model, the single GRU model, and the TCN-GRU model without denoising, the TCN-GRU model integrated with optimized EWMA denoising achieves better prediction performance on the test set. In terms of the early warning matching rate, it reaches 83.33% for the t + 1 horizon, and the prediction accuracy for the t + 5 and t + 10 horizons decreases regularly but remains stable above 60%. In contrast, the highest early warning matching rate of the model without denoising is only 22.22% across all horizons, which has no practical early warning value. The early warning signals generated by the optimized EWMA denoising-based TCN-GRU model can effectively identify abnormal sharp rises and falls in egg futures prices, providing effective support for hedging and risk management for market participants. The study’s limitations are discussed, as well as future research directions. The findings provide a basis for decision making for agricultural producers and future investors and support the development of China’s agricultural product market. Full article

Short-term traffic load prediction is a fundamental component of intelligent transportation systems (ITSs), supporting real-time monitoring, congestion mitigation, and adaptive traffic management in smart cities. Owing to the dynamic and nonlinear nature of urban traffic, identifying prediction models that align with real-world traffic dynamics remains a key challenge. This study presents a comparative evaluation of data-driven traffic load prediction models using high-resolution one-minute traffic data collected from a major urban roundabout in Jeddah, Saudi Arabia. The evaluated models include regression-based machine learning approaches and recurrent deep learning architectures, which are assessed under consistent preprocessing and evaluation conditions. Model performance is evaluated using standard error metrics and complemented by temporal and residual analyses to examine prediction behavior under different traffic regimes. The optimized GRU model achieved the best predictive accuracy with an RMSE of 149.12 veh/h, followed closely by the optimized LSTM model (RMSE = 150.85 veh/h). The results indicate that while conventional machine learning models can effectively capture overall traffic trends under relatively stable conditions, recurrent deep learning models demonstrate stronger capability in modeling nonlinear temporal dependencies and rapid traffic fluctuations when properly configured. In addition, a variability-based regime analysis was conducted to evaluate model robustness under different traffic demand dynamics, revealing that model performance advantages are context-dependent rather than universal. The findings highlight the importance of systematic comparative evaluation and data-driven model selection for developing reliable traffic prediction components in real-time ITS applications and sustainable urban mobility planning. Full article

Accurate prediction of embankment settlement and evaluation of its serviceability in permafrost regions are significantly challenged by scarce monitoring data and dynamic, non-stationary settlement processes. To address this, an integrated framework combining change-point detection with a novel dynamic prediction model is proposed. Analysis of long-term monitoring data from the Qinghai–Tibet Railway using the Pettitt test revealed a key change point around 2015, indicating a transition towards stabilization. Subsequently, an SAA-GRU-LSTM hybrid model, employing a dynamic compensation prediction strategy, was developed. The model successfully utilized only early-stage data to forecast future settlement trends, demonstrating robust performance in adapting to the identified abrupt change. Furthermore, by applying established engineering serviceability criteria to both historical and predicted data, the framework enables a dynamic and prospective serviceability assessment. This methodology provides a practical tool for the maintenance and risk management of infrastructure in permafrost environments under conditions of data scarcity and process uncertainty. Full article