Search Results (179)

This article presents a comprehensive review of methods used for forecasting electricity consumption. The studies analyzed by the authors encompass both classical statistical models and modern approaches based on artificial intelligence, including machine-learning and deep-learning techniques. Electricity load forecasting is categorized into four time horizons: very short term, short term, medium term, and long term. The authors conducted a comparative analysis of various models, such as autoregressive models, neural networks, fuzzy logic systems, hybrid models, and evolutionary algorithms. Particular attention was paid to the effectiveness of these methods in the context of variable input data, such as weather conditions, seasonal fluctuations, and changes in energy consumption patterns. The article emphasizes the growing importance of accurate forecasts in the context of the energy transition, integration of renewable energy sources, and the management of the evolving electricity system, shaped by decentralization, renewable integration, and data-intensive forecasting demands. In conclusion, the authors highlight the lack of a universal forecasting approach and the need for further research on hybrid models that combine interpretability with high predictive accuracy. This review can serve as a valuable resource for decision-makers, grid operators, and researchers involved in energy system planning. Full article

This study proposes a hybrid approach to analyze and forecast non-stationary financial time series by combining statistical models with deep neural networks. A model is introduced that integrates three key components: the Generalized Autoregressive Score (GAS) model, which captures volatility dynamics; an attention mechanism (ATT), which identifies the most relevant features within the sequence; and a Long Short-Term Memory (LSTM) neural network, which receives the outputs of the previous modules to generate price forecasts. This architecture is referred to as GAS-ATT-LSTM. Both unidirectional and bidirectional variants were evaluated using real financial data from the Nasdaq Composite Index, Invesco QQQ Trust, ProShares UltraPro QQQ, Bitcoin, and gold and silver futures. The proposed model’s performance was compared against five benchmark architectures: LSTM Bidirectional, GARCH-LSTM Bidirectional, ATT-LSTM, GAS-LSTM, and GAS-LSTM Bidirectional, under sliding windows of 3, 5, and 7 days. The results show that GAS-ATT-LSTM, particularly in its bidirectional form, consistently outperforms the benchmark models across most assets and forecasting horizons. It stands out for its adaptability to varying volatility levels and temporal structures, achieving significant improvements in both accuracy and stability. These findings confirm the effectiveness of the proposed hybrid model as a robust tool for forecasting complex financial time series. Full article

Indonesia’s dependence on fossil fuel exports, particularly coal and crude oil, presents a dual challenge: sustaining economic growth while addressing rising CO₂ emissions. Despite significant attention to domestic energy consumption, the environmental implications of export activities remain underexplored. This study examines the dynamic relationship between energy exports, crude oil prices, and CO₂ emissions in Indonesia using a Vector Autoregressive (VAR) model with annual data from 2002 to 2022. The analysis incorporates Impulse Response Functions (IRFs) and Forecast Error Variance Decomposition (FEVD) to trace short- and long-term interactions among variables. Findings reveal that coal exports are strongly persistent and positively linked to past emission levels, while oil exports respond negatively to both coal and emission shocks—suggesting internal trade-offs. CO₂ emissions are primarily self-driven yet increasingly influenced by oil export fluctuations over time. Crude oil prices, in contrast, have limited impact on domestic emissions. This study contributes a novel export-based perspective to Indonesia’s emission profile and demonstrates the value of dynamic modeling in policy analysis. Results underscore the importance of integrated strategies that balance trade objectives with climate commitments, offering evidence-based insights for refining Indonesia’s nationally determined contributions (NDCs) and sustainable energy policies. Full article

Accurate gas consumption forecasting is critical for modern energy systems due to complex consumer behavior and regulatory requirements. Deep neural networks (DNNs), such as Seq2Seq with attention, TiDE, and Temporal Fusion Transformers, are promising for modeling complex temporal relationships and non-linear dependencies. This study compares state-of-the-art architectures using real-world data from over 100,000 consumers to determine their practical viability for forecasting gas consumption under operational and regulatory conditions. Particular attention is paid to the impact of data quality, feature attribution, and model reliability on performance. The main use cases for natural gas consumption forecasting are tariff setting by regulators and system balancing for suppliers and operators. The study used monthly natural gas consumption data from 105,527 households in the Volyn region of Ukraine from January 2019 to April 2023 and meteorological data on average monthly air temperature. Missing values were replaced with zeros or imputed using seasonal imputation and the K-nearest neighbors. The results showed that previous consumption is the dominant feature for all models, confirming their autoregressive origin and the high importance of historical data. Temperature and category were identified as supporting features. Improvised data consistently improved the performance of all models. Seq2SeqPlus showed high accuracy, TiDE was the most stable, and TFT offered flexibility and interpretability. Implementing these models requires careful integration with data management, regulatory frameworks, and operational workflows. Full article

Serverless computing has gained significant attention in recent years. However, the cold start problem remains a major challenge, not only because of the substantial latency it introduces to function execution time, but also because frequent cold starts lead to poor resource utilization, especially during workload fluctuations. To address these issues, we propose a multi-level scheduling solution: the Hybrid Model. This model is designed to reduce the frequency of cold starts while maximizing container utilization. At the global level (across invokers), the Hybrid Model employs a skewness-aware scheduling strategy to select the most appropriate invoker for each request. Within each invoker, we introduce a greedy buffer-aware scheduling method that leverages the available slack (remaining buffer) of warm containers to aggressively encourage their reuse. Both the global and the local schedule are tightly integrated with a prediction component- The Hybrid Predictor- that combines Auto-Regressive Integrated Moving Average ARIMA (linear trends) and Random Forest (non-linear residuals + environment-aware features) for 5-min workload forecasts. The Hybrid Model is implemented on Apache OpenWhisk and evaluated using Azure-like traces and real FaaS applications. The evaluations show that the Hybrid Model achieves up to 34% SLA violation reductions compared to three state-of-the-art approaches and maintains the container utilization to be more than 80%. Full article

As the relationship between cryptocurrency mining activities and electricity consumption becomes increasingly close, the risk spillover effect is steadily drawing a lot of attention to the energy and cryptocurrency markets. For the purpose of studying the risk contagion between the cryptocurrency and energy market, this paper constructs a risk contagion network between cryptocurrency and China’s energy market using complex network methods. The tail risk spillover effects under various time and frequency domains were captured by the spillover index, which was assessed by the leptokurtic quantile vector autoregression (QVAR) model. Considering the spatial heterogeneity of energy companies, the spatial Durbin model was used to explore the impact mechanism of risk spillovers. The research showed that the framework of this paper more accurately reflects the tail risk spillover effect between China’s energy market and cryptocurrency market under various shock scales, with the extreme state experiencing a much higher spillover effect than the normal state. Furthermore, this study found that the tail risk contagion between cryptocurrency and China’s energy market exhibits notable dynamic variation and cyclical features, and the long-term risk spillover effect is primarily responsible for the total spillover. At the same time, the study found that the company with the most significant spillover effect does not necessarily have the largest company size, and other factors, such as geographical location and business composition, need to be considered. Moreover, there are spatial spillover effects among listed energy companies, and the connectedness between cryptocurrency and the energy market network generates an obvious impact on risk spillover effects. The research conclusions have an important role in preventing cross-contagion of risks between cryptocurrency and the energy market. Full article

As Advanced Persistent Threats (APTs) continue to evolve, constructing a dynamic cybersecurity knowledge graph requires precise extraction of entity–relationship triples from unstructured threat intelligence. Existing approaches, however, face significant challenges in modeling low-frequency threat associations, extracting multi-relational entities, and resolving overlapping entity scenarios. To overcome these limitations, we propose the Symmetry-Aware Prototype Contrastive Learning (SAPCL) framework for joint entity and relation extraction. By explicitly modeling syntactic symmetry in attack-chain dependency structures and its interaction with asymmetric adversarial semantics, SAPCL integrates dependency relation types with contextual features using a type-enhanced Graph Attention Network. This symmetry–asymmetry fusion facilitates a more effective extraction of multi-relational triples. Furthermore, we introduce a triple prototype contrastive learning mechanism that enhances the robustness of low-frequency relations through hierarchical semantic alignment and adaptive prototype updates. A non-autoregressive decoding architecture is also employed to globally generate multi-relational triples while mitigating semantic ambiguities. SAPCL was evaluated on three publicly available CTI datasets: HACKER, ACTI, and LADDER. It achieved F1-scores of 56.63%, 60.21%, and 53.65%, respectively. Notably, SAPCL demonstrated a substantial improvement of 14.5 percentage points on the HACKER dataset, validating its effectiveness in real-world cyber threat extraction scenarios. By synergizing syntactic–semantic multi-feature fusion with symmetry-driven dynamic representation learning, SAPCL establishes a symmetry–asymmetry adaptive paradigm for cybersecurity knowledge graph construction, thus enhancing APT attack tracing, threat hunting, and proactive cyber defense. Full article

Reverberation is the primary interference of active detection. Therefore, the effective suppression of reverberation is a prerequisite for reliable signal processing. Existing dereverberation methods have shown effectiveness in specific scenarios. However, they often struggle to exploit the distinction between target echo and reverberation, especially in complex, dynamically changing underwater environments. This paper proposes a novel dereverberation network, ERCL-AttentionNet (Echo–Reverberation Complementary Learning Attention Network). We use the Continuous Wavelet Transform (CWT) to extract time–frequency features from the received signal, effectively balancing the time and frequency resolution. The real and imaginary parts of the time–frequency matrix are combined to generate attention representations, which are processed by the network. The network architecture consists of two complementary UNet models sharing the same encoder. These models independently learn target echo and reverberation features to reconstruct the target echo. An attention mechanism further enhances performance by focusing on target information and suppressing irrelevant disturbances in complex environments. Experimental results demonstrate that our method achieves a higher Peak-to-Average Signal-to-Reverberation Ratio (PSRR), Structural Similarity Index (SSIM), and Peak-to-Average Ratio (PAR) of cross-correlation while effectively preserving key time–frequency features, compared to traditional methods such as autoregressive (AR) and singular value decomposition (SVD). Full article

Radar echo extrapolation, a critical spatiotemporal sequence forecasting task, requires precise modeling of motion trajectories and intensity evolution from sequential radar reflectivity inputs. Contemporary deep learning implementations face two operational limitations: progressive attenuation of predicted echo intensities during autoregressive inference and spectral leakage-induced diffusion at high-intensity echo boundaries. This study presents RaDiT, a hybrid architecture combining differential transformer with adversarial training for radar echo extrapolation. The framework employs a U-Net backbone augmented with vision transformer blocks, utilizing differential attention mechanisms to govern spatiotemporal interactions. Our differential attention mechanism enhances noise suppression under high-threshold conditions, effectively minimizing spurious feature generation while improving metric reliability. A conditional GAN discriminator is integrated to maintain microphysical consistency in generated sequences, simultaneously addressing spectral blurring and intensity dissipation. Comprehensive evaluations demonstrate RaDiT’s superior performance in preserving spatiotemporal coherence and intensity across 0–90 min forecasting horizons. The proposed architecture achieves CSI improvements of 10.23% and 2.88% at 4 × 4 and 16 × 16 spatial pooling scales, respectively, for ≥30 dBZ thresholds on the CMARC dataset compared to PreDiff. To our knowledge, this represents the first successful implementation of differential transformers for radar echo extrapolation. Full article

In recent years, scholars have been paying more attention to financial inclusion, which has been positioned as a crucial component in accomplishing the majority of the seventeen Sustainable Development Goals set forward by the United Nations. Investigating the effects of life insurance and financial inclusion on poverty in 45 Sub-Saharan African (SSA) nations between 1999 and 2023 is the goal of this study. Using the Panel Autoregressive Distributed Lag (P-ARDL) method, this study concludes that poverty can be decreased through financial inclusion. Notably, we found that life insurance raises poverty when financial inclusion follows. This might be because there are not many microinsurance options available in SSA nations for those with low incomes. Due to their increased likelihood of being financially illiterate and their inability to purchase the necessary smart devices and internet services, the lower-income segments are unable to enjoy the same advantages as the higher-income segments. According to the findings, financial exclusion problems may be resolved by future life insurance, but this must be done in a sustainable manner. Future life insurance should address the requirements of the underprivileged and lower-income groups, and financial inclusion should be progressively enhanced. Full article

The rising global demand for food and energy has led to growing attention to the nexus between food security and renewable energy. This study aims to investigate the impacts and shocks of renewable energy consumption, particularly solar and wind energy, on food availability and stability in Saudi Arabia, by assessing both short-term and long-term effects. We use the time series annual data covering the period (2000–2022) analyzed by applying the Vector Autoregressive (VAR) model system and its environment, Granger causality, the forecast-error variance decompositions (FEVD), and the impulse response functions (IRFs). The VAR results indicated that wind renewable energy positively affects food availability; one unit of wind energy consumption will significantly increase food availability by 3.16% (Z value 2.017 at a 5% significance level), and no statistically significant coefficients are associated with food stability. Also, the results confirmed that one unit of renewable energy consumption from solar will significantly increase food stability by 36.5% in Saudi Arabia (Z-value 1.682 at a 10% significance level). The Granger causality results concluded that solar energy has a bidirectional Granger causality with food availability but not food stability. The FEVD results showed that solar energy shocks have more persistent impacts in explaining the rapid increase in food security than wind energy shocks in both the short and long term. The IRFs concluded that food availability has shown a positive and steady increase in response to wind energy. This study provides practical recommendations for policymakers to balance energy transition goals with food security concerns. Future research should explore emerging technologies in wind and solar energy that can enhance efficiency and sustainability while minimizing adverse effects on food security. Full article

Ocean forecasting is critical for various applications and is essential for understanding air–sea interactions, which contribute to mitigating the impacts of extreme events. While data-driven forecasting models have demonstrated considerable potential and speed, they often primarily focus on spatial variations while neglecting temporal dynamics. This paper presents the TSformer, a novel non-autoregressive spatio-temporal transformer designed for medium-range ocean eddy-resolving forecasting, enabling forecasts of up to 30 days in advance. We introduce an innovative hierarchical U-Net encoder–decoder architecture based on 3D Swin Transformer blocks, which extends the scope of local attention computation from spatial to spatio-temporal contexts to reduce accumulation errors. The TSformer is trained on 28 years of homogeneous, high-dimensional 3D ocean reanalysis datasets, supplemented by three 2D remote sensing datasets for surface forcing. Based on the near-real-time operational forecast results from 2023, comparative performance assessments against in situ profiles and satellite observation data indicate that the TSformer exhibits forecast performance comparable to leading numerical ocean forecasting models while being orders of magnitude faster. Unlike autoregressive models, the TSformer maintains 3D consistency in physical motion, ensuring long-term coherence and stability. Furthermore, the TSformer model, which incorporates surface auxiliary observational data, effectively simulates the vertical cooling and mixing effects induced by Super Typhoon Saola. Full article

High-resolution remote sensing imagery has become the primary data source for obtaining building information. Automatically extracting regularized building outline polygon vectors is crucial for improving vector mapping efficiency and geographic information system applications, but existing deep learning methods struggle to simultaneously achieve accurate detection, high pixel-level coverage, and geometric regularity. This paper proposes a novel two-stage building outline extraction method. In the first stage, the SegFormer model is used to extract image features, effectively capturing global context information. In the second stage, a polygon outline regularization model (PolyReg) based on a Masked Attention Encoder is innovatively introduced. The PolyReg model draws on the sequence generation idea from natural language processing, transforming the outline regularization task into a sequence generation problem. Through a cleverly designed self-attention mask matrix, it achieves an autoregressive output of regularized building outline coordinates, eliminating the need for cumbersome post-processing steps. Experimental results show that on the Inria Aerial Image Labeling Dataset, compared with traditional methods and existing deep learning methods, the proposed method demonstrates significant advantages in metrics such as IoU, C-IoU, and Hausdorff distance. It effectively improves the regularity and geometric accuracy of building outlines while maintaining high pixel-level coverage. Full article

As the global energy transition progresses, green electricity, which is crucial for low-carbon systems, has gained attention. However, the lack of effective market linkages hinders a full understanding of the price transmission effects across green markets. This study uses the Vector Autoregression (VAR) model and Granger causality tests to analyze the price transmission and lag effects between the carbon, green certificate, and China Certified Emission Reduction (CCER) Markets. The findings reveal complex price linkages, offering theoretical insights and policy recommendations for optimizing green electricity markets and environmental rights trading. Full article

In recent years, deep-learning-based speech synthesis has garnered substantial attention, achieving remarkable advancements in generating human-like speech. However, synthesized speech often lacks naturalness, primarily because models excessively depend on fine-grained text–speech alignment. To address this issue, we propose MixDiff-TTS, a novel non-autoregressive model. MixDiff-TTS incorporates a linguistic encoder based on a mixture alignment mechanism, which combines word-level hard alignment with phoneme-level soft alignment. This design reduces reliance on fine-grained alignment, enabling the model to handle ambiguous phonetic boundaries more robustly. Additionally, we introduce a Word-to-Phoneme Attention module with a relative position bias mechanism to improve the model’s capacity for processing long text sequences. We evaluate the performance of MixDiff-TTS on the LJSpeech dataset. The experimental results show that MixDiff-TTS scores 0.507 for SSIM (Structural Similarity Index) and 6.652 for MCD (Mel Cepstral Distortion). This suggests that the synthesized speech is closer to real speech in spectral structure and exhibits lower spectral distortion than state-of-the-art baselines (such as FastSpeech2 and DiffSpeech). MixDiff-TTS also achieves a MOS (Mean Opinion Score) of 3.95, which is close to that of real speech. These results indicate that MixDiff-TTS can synthesize speech with high naturalness and quality. Ablation studies demonstrate the effectiveness of our method. Full article