Search Results (706)

To address the insufficient accuracy of short-term load forecasting caused by the strong randomness of distributed energy output, variable electricity consumption patterns, and complex meteorological factors, this study proposes a load forecasting method that integrates K-means scene clustering and a Transformer–BiGRU–Attention (CTBA) hybrid deep learning architecture. Different from conventional Transformer–BiGRU hybrid forecasters that train a single global predictor across all operating conditions, the proposed CTBA framework first partitions daily load curves into representative scenes and then routes each sample to a scene-specific Transformer–BiGRU–Attention predictor, thereby reducing distributional heterogeneity before temporal modeling. First, the K-means algorithm is used to perform scene clustering on historical daily load curves, and the optimal number of clusters is selected according to the silhouette coefficient and downstream prediction performance. Subsequently, the CTBA model is trained separately for each clustering subset. The Transformer encoder captures the long-range global dependencies of load sequences through the self-attention mechanism, the BiGRU module extracts local bidirectional temporal fluctuation features, and the Attention mechanism further focuses on key time nodes such as morning and evening peaks while fusing multi-source data including historical load, day-ahead electricity price, and multi-dimensional meteorological factors. Experimental results based on the German ENTSO-E power dataset show that the coefficient of determination R² of the proposed model reaches 0.9893, with MAE, RMSE, and MAPE as low as 0.0141, 0.0187, and 3.92%, respectively, which are significantly improved compared to benchmark models such as SVR, LSTM, CNN, and TCN-BiGRU. Ablation experiments further demonstrate that removing the clustering, Transformer, BiGRU, or attention layer will degrade performance, thus verifying the effectiveness and superiority of the method in short-term load forecasting and providing an accurate solution for the short-term load forecasting of power systems. Full article

The efficient management of power systems requires balancing electricity generation costs with associated environmental emissions under dynamically varying demand. This paper proposes a two-stage approach that combines machine learning (ML) with a metaheuristic optimization algorithm to address the dynamic economic–environmental load dispatch (DEELD) challenge. In the first stage, electricity consumption data are enriched with temporal features to capture demand patterns and enable accurate forecasting. In the second stage, the daily scheduling horizon is divided into multiple periods, and dispatch solutions are generated sequentially while enforcing ramp-rate constraints. To enhance operational realism, a priority-based generator scheduling mechanism is explicitly introduced, enforcing hierarchical unit commitment and reflecting practical dispatch policies. Rather than focusing on a single optimal solution, the proposed framework generates multiple feasible dispatch solutions and evaluates them using economic, environmental, and operational performance indicators. These solutions are then ranked according to predefined decision profiles, enabling system operators to select dispatch strategies that align with specific priorities. This transforms the dispatch process into a flexible decision-support tool capable of addressing diverse real-world requirements. Full article

Residential energy consumption exhibits substantial behavioral uncertainty and temporal heterogeneity, which pose challenges for demand-side management and residential load profiling. However, existing studies often focus on isolated temporal or spatial scales and predominantly employ hard clustering methods based on geometric distance metrics. To address these limitations, this study proposes a multi-scale residential load profiling framework utilizing the Gaussian Mixture Model (GMM) and nearly three years of hourly electricity consumption data from 13 residential buildings in Vancouver. First, schedule-driven and seasonal variations in residential energy consumption were examined through multi-temporal comparative analyses and paired-sample t-tests. The results indicate statistically significant differences between working-time and non-working-time energy consumption patterns in most buildings (p < 0.001). Second, individual-building clustering was performed to identify long-term intra-building daily load evolution characteristics, revealing 2–5 typical daily profiles across different households. Finally, inter-building clustering identified three representative residential groups characterized by low-energy stable patterns, high-energy intensive patterns, and intermediate commuting-oriented patterns. The average daily energy consumption levels of the three clusters were 13.11 kWh, 36.74 kWh, and 21.61 kWh, respectively. The proposed framework provides a data-driven approach for understanding residential energy-use heterogeneity across multiple scales and offers potential guidance for residential demand-side management and urban low-carbon energy planning. Full article

Against the backdrop of global decarbonization in energy-intensive industries, the primary aluminum sector has become a critical field for deep industrial decarbonization due to its high electricity consumption, large share of indirect carbon emissions, and complex mitigation pathways. This challenge is particularly salient in regions endowed with abundant renewable resources while hosting concentrated industrial electricity demand, where coordinated mitigation across technological upgrading and energy system transformation has broad practical relevance. Using Xining in Qinghai Province, China, a renewable-rich region, as an illustrative case, this study systematically examines the major carbon mitigation pathways in the primary aluminum industry, including mining, alumina production, electrolytic cell retrofitting, power system coordination, and carbon capture, utilization, and storage (CCUS). A multi-objective optimization model is developed to minimize marginal abatement costs (MAC) while maximizing technological application performance, and the sequential unconstrained minimization technique (SUMT) is employed to optimize mitigation pathways under short-, medium-, and long-term scenarios. The results show that, in the short term (before 2030), emission reduction mainly relies on improvements in electrolysis efficiency, leading to a mitigation pattern dominated by reductions in electricity consumption per unit of output. In the medium term (before 2035), the pathway shifts from isolated process optimization to a coordinated strategy combining process upgrading with power decarbonization, exhibiting a structural mitigation pattern driven by synergy between the production side and the energy side. In the long term (before 2060), the pathway evolves toward a stage dominated by energy system reconfiguration and carbon utilization. With high shares of renewable electricity integration, DC power supply configurations, and energy storage support, primary aluminum production is expected to achieve deep decarbonization on the power side. This study provides a transferable analytical framework and policy-relevant insights for the low-carbon transition of energy-intensive industries in renewable-rich regions. Full article

Accurate customer baseline load (CBL) estimation is crucial for incentive allocation and flexibility potential assessment in demand response (DR) programs. However, residential electricity consumption is highly stochastic, and long-duration DR events often result in missing critical load segments, making it difficult for traditional regression-based and daily load-profile clustering methods to accurately capture the counterfactual baseline pattern. To address this issue, this paper proposes a CBL estimation method that integrates a physics-/domain-informed response-consistency constraint with a conditional generative adversarial network. In the proposed framework, deep soft clustering is employed to extract weekly scale load modes, while mutual information (MI) and autocorrelation coefficient (ACC) are quantified as user-specific conditioning fingerprints to characterize intrinsic consumption behaviors. Comparative experiments on a publicly available real-world dataset demonstrate that the proposed method provides strong event-period accuracy among the recurrent and attention-based benchmark models considered in the main comparison. Under matched response-consistency budgets, PI-ICGAN achieves the lowest constrained DR-period MAE at the tested NRR targets, and the ablation results show that the attention, fingerprint, response-consistency, and GradNorm components contribute to different aspects of the accuracy–consistency trade-off. Full article

Cities preparing Sustainable Urban Mobility Plans (SUMPs) increasingly require practical tools capable of merging diverse mobility datasets and transforming them into planning-relevant indicators. This article introduces PRIORITY (Platform for the tRansition to sustaInable zerO-caRbon mobilITY), a prototype platform designed to support mobility analysis and decision-making in urban contexts. The platform integrates Floating Car Data, GTFS feeds describing public transport supply, and detailed land-use and zoning information. By relying on these heterogeneous data streams, PRIORITY generates indicators such as travel and stop times, trip distances, trip volumes, energy consumption, pollutant emissions, external costs, and electric-vehicle charging behavior. The platform is organized into two main components: a back end and a front end. The back end, which constitutes the operational core, manages all collected data and ensures their structured storage in a shared database capable of handling large volumes of information on urban form, individual mobility patterns, public transport services, and modeling outcomes. The front end provides an intuitive and versatile interface that dynamically presents the outputs generated by the platform’s analytical and modeling processes. A case application for the Metropolitan City of Rome (Italy) illustrates the operational use of the prototype and shows how PRIORITY can support transparent and reproducible evaluations during the preparation and monitoring of SUMPs. The demonstrated workflow highlights the prototype’s value for public authorities and planners seeking data-informed approaches to urban mobility assessment and decarbonization strategies. Full article

The growing volatility and complexity of global food supply chains have intensified the need for integrated analytical frameworks capable of supporting anticipatory and data-driven decision-making. This article examines how logistics sector observatories can function as strategic intelligence infrastructures for identifying structural tensions and supporting resilience in cold-chain logistics systems. The article introduces the concept of logistics sector observatories as strategic intelligence infrastructures and examines its empirical relevance through a longitudinal analysis of the Spanish cold-chain logistics sector. Empirically, the research draws on a multi-source dataset constructed through the ALDEFE Observatory in collaboration with industry stakeholders over the core study period 2021–2025, encompassing storage capacity, consumption dynamics, energy costs, international logistics indices, and macroeconomic variables. Complementary energy benchmark data for 2019–2025 are used to contextualize electricity cost volatility. Methodologically, the study combines qualitative insights from stakeholder interviews with exploratory quantitative longitudinal analysis. The results suggest severe structural tensions driven by the interaction between rigid capacity constraints and energy cost volatility. The analysis identifies a pattern of persistently high storage occupancy despite substantial energy-price fluctuations. This finding is consistent with the structural inelasticity of cold-chain demand, which reduces operational slack and affects system resilience. Beyond operational resilience, the study highlights the potential contribution of sector observatories to the energy sustainability transition through future sector-level indicators related to energy intensity, refrigeration efficiency, and carbon performance. The study contributes a sector-level, data-driven perspective on visibility, coordination, and anticipatory governance in complex logistics environments. Full article

Predictive energy management strategies (PEMSs) have attracted increasing attention in hybrid electric vehicles (HEVs) for improving fuel economy and powertrain efficiency using anticipated driving information. For PEMS, data-driven velocity prediction is widely used to capture complex driving patterns from historical trajectories and future traffic priors, but often lacks kinematic awareness, leading to physical causality violations and long-horizon state drift. To address these issues, this paper proposes a physics-informed PEMS, where a Physics-Informed Spatio-Temporal Network (PI-STN) provides control-oriented velocity information for an MPC-based energy management controller. Specifically, to address pseudo-motion in velocity prediction under standstill conditions, a global zero-speed gating mechanism is introduced; to suppress acceleration/deceleration trends that violate vehicle kinematic causality, a causal penalty is designed; and to mitigate temporal phase misalignment between data-driven predictions and physical motion priors, a Differentiable Kalman Filter (DKF) is incorporated. At each receding horizon step, the PI-STN-predicted velocity sequence is converted into future power demand through longitudinal vehicle dynamics and used by MPC for engine–battery power allocation under SOC and engine transient constraints. Under the same tested conditions, the proposed strategy reduces engine power fluctuation by 15.1% compared with BiLSTM-Transformer, and achieves an equivalent fuel consumption of 323.74 g, outperforming Transformer-KF by 3.12%. Full article

Electricity theft remains a critical challenge that destabilizes power systems, causes significant financial losses, and disrupts the grid, particularly in developing countries. This study presents a machine learning framework integrating an ANN and advanced performance metrics to accurately detect fraud consumption patterns in a smart meter dataset. The method achieves strong categorization between normal and abnormal conduct by simulating temporal behavior across seasons, applying feature extraction to high-resolution energy signals, and assessing performance using RMSE, MAE, and R². The experimental results demonstrate that intelligent algorithms significantly improve theft-detection accuracy; reduce losses, especially NTLs; and provide a scalable foundation for future smart-grid security. Full article

The reduction in driving range during winter remains a major barrier to the widespread adoption of battery electric buses (BEBs), as battery performance degradation and increased Heating, Ventilation and Air Conditioning (HVAC) energy demand significantly raise total energy consumption. This study investigates the use of proton exchange membrane fuel cells (PEMFCs) as auxiliary power units for range-extended electric buses (FC-REEBs) under low-temperature conditions to address this challenge. A comprehensive dynamic model was developed in MATLAB/Simulink 2025a version, integrating a fuel cell system, lithium-ion battery, power conversion unit, vehicle dynamics, and cabin thermal model. The model was evaluated under the World Harmonized Vehicle Cycle (WHVC) to compare three fuel cell operation strategies defined by fuel cell capacity and operating modes for cabin heating and battery charging. Performance was compared in terms of SOC variation, fuel cell loading patterns, hydrogen consumption, and equivalent fuel economy. Results indicate that the high-capacity strategy improves SOC stability but increases hydrogen consumption and reduces overall efficiency. In contrast, the strategy prioritizing cabin heating with minimal battery charging effectively utilizes waste heat and achieves the highest equivalent fuel economy. These findings highlight key trade-offs among different operating strategies and demonstrate that fuel cells can significantly enhance BEB efficiency and driving performance in cold environments while reducing battery load. Full article

The growing emphasis on environmental sustainability within the European Union raises important questions about the internal structure of corporate environmental effort. This study examines environmental expenditures (intermediate consumption of environmental protection services) and environmental investments (gross fixed capital formation for environmental protection) in ten EU member states over 2015–2022, using Eurostat Environmental Protection Expenditure Accounts data at both the national and sectoral levels (NACE Rev.2 sectors A, B, C, D). Two hypotheses are tested empirically. First, sectoral differences in the investment-to-expenditure ratio are statistically significant (Kruskal–Wallis H = 27.72, p < 0.0001): electricity, mining, and manufacturing each display a higher ratio than agriculture, with the most pronounced contrast for electricity. Second, Eastern European member states exhibit a systematically higher investment-to-expenditure ratio than the remaining countries in the sample (level difference: β = +1.01, p < 0.001), although the two groups follow parallel trajectories without convergence or divergence over the period examined. Building on the relative intensity of the two indicators, the study proposes a four-quadrant typology—active transformation, investment focus, maintenance model, and low-intensity profile—whose stability is confirmed by bootstrap resampling, sub-sample analysis, and an alternative deflator specification. The findings suggest that the internal composition of environmental effort is as informative as its overall level and that sectoral disaggregation is essential for characterising patterns of environmental effort in the EU. Full article

The emergence of citizen energy communities under the European Clean Energy Package is creating new opportunities for neighboring households to collectively reduce electricity costs through local energy sharing. This paper presents a distributed multi-agent energy management system for a two-household residential energy community in which each household is equipped with photovoltaic generation and a battery energy storage system operating under realistic hourly-varying electricity prices. Each household is managed by an independent Deep Q-Learning agent that learns a cost-optimal charging and discharging policy using only local observations. In parallel, a coordination agent, implemented on the SPADE platform with XMPP-based messaging, oversees real-time peer-to-peer energy transfers between households, enabling energy exchange whenever one household has surplus generation and another faces a deficit. The two households are deliberately configured with complementary profiles: one has higher PV generation capacity while the other has higher energy consumption. This setup creates natural opportunities for local energy sharing between them. Performance is assessed through a three-level evaluation framework: (i) individual household economics (cost reduction, battery management, grid exchanges), (ii) coordination efficiency (transfer frequency, direction, and volume), and (iii) aggregate community performance, which isolates the added value of peer-to-peer sharing beyond what each household achieves through individual BESS optimization. Numerical experiments using GEFCom2014 solar generation data, synthetic residential load profiles calibrated following documented consumption patterns, and day-ahead price signals representative of the Spanish electricity market demonstrate that both Deep Q-Learning agents independently learn effective charge/discharge strategies aligned with price signals and PV availability. They also show that the coordination layer further reduces community grid dependence by routing surplus energy locally rather than exchanging it with the main grid at less favorable rates. The results confirm that a well-engineered integration of decentralized reinforcement learning with a lightweight coordination protocol can deliver measurable economic benefits in realistic residential energy communities without requiring centralized training, shared data, or complex multi-agent reinforcement learning architectures. Full article

Hydrogen energy, owing to its advantages of low-carbon cleanliness, long-term storage capacity, and multi-energy coupling potential, has emerged as a crucial medium for enhancing renewable energy accommodation within integrated energy systems. However, the pronounced heterogeneity in hydrogen load behaviors, temporal characteristics, and regulation capabilities poses significant challenges for unified modeling approaches, which struggle to accurately capture the multi-modal regulation potential of hydrogen demand, thereby limiting the precision of system operation optimization. To address this issue, this paper proposes a data-driven hydrogen load flexibility modeling method for integrated energy system (IES) operation optimization. A hybrid LSTM-ISODATA framework is designed to extract deep temporal dependencies and identify six representative hydrogen consumption patterns from typical load sequences. Each hydrogen load category is decomposed into shiftable, transferable, and reducible flexible forms, and a category-specific time-varying flexibility constraint matrix is established to characterize differentiated regulation capabilities. An electricity–heat–hydrogen integrated energy system operation optimization model embedded with classified flexible hydrogen loads is developed and solved via mathematical programming. Simulation results show that the proposed method reduces system operating costs by 10.3% compared with conventional unified modeling, while significantly promoting renewable energy utilization and system operational flexibility. The effectiveness and engineering applicability of the proposed model in IES optimal scheduling are fully validated. Full article

Achieving carbon neutrality, enhancing energy efficiency, securing energy supply, and accurately forecasting energy demand are among the most urgent global energy priorities. In this study, Turkey’s geothermal, wind, and solar electricity consumption was forecasted for the 2025–2030 period using five years of historical data through eight different regression-based models. The forecast models included ARIMA, Linear Regression, Polynomial Regression, Exponential Smoothing, Ridge, Lasso, SVR, and XGBoost. Forecast accuracy was validated using 2023–2024 data. A hybrid model, integrating the Lasso and Random Forest approaches via weighted averaging, was developed to enhance forecast robustness. Pearson correlation was applied to quantify the impact of key socioeconomic variables—such as population, GDP, and university graduates—on energy consumption patterns. Forecast comparisons revealed that Random Forest and XGBoost produced results closest to the Hybrid model, with deviation rates of 1.84–7.27% and 0.03–1.08%, respectively. In contrast, Polynomial Regression and Exponential Smoothing showed significant biases, with deviations reaching up to 61.58% and 54.48% in 2030. ARIMA remained relatively consistent but exhibited increasing deviation over time. The Exponential and Polynomial models consistently overestimated demand, while SVR underestimated it throughout the forecast horizon. Ridge Regression provided stable but systematically higher forecasts. The findings indicate that the hybrid model provides a balanced forecasting structure and mitigates the under- or overestimation tendencies observed in singular models. This research supports strategic, data-driven energy planning in alignment with long-term sustainability goals. Full article

The rapid electrification of road transport has increased interest in distributed energy strategies that reduce grid demand and support decarbonization. Vehicle-integrated photovoltaics (VIPV), including vehicle-applied photovoltaic configurations (VAPV), can generate electricity directly on the vehicle. This systematic review examines peer-reviewed VIPV literature published between 2015 and 2026, focusing on the distinction between theoretical photovoltaic generation and practically usable energy. A Scopus search conducted on 2 May 2026 identified 196 records, of which 88 studies were included after screening against predefined criteria. Due to heterogeneity in vehicle types, climates, technologies, modeling assumptions, and reported metrics, no meta-analysis was performed. Instead, the review applies a multi-layered framework covering climate, geometry, thermal effects, electrical mismatch, battery state-of-charge interactions, fleet-scale modeling, economics, and life-cycle implications. The evidence shows that VIPV is technically feasible and can deliver measurable energy yields, especially in high-irradiance regions and vehicles with favorable daytime parking exposure. However, useful contribution depends strongly on curvature losses, dynamic shading, electrical configuration, SOC limits, charging behavior, seasonality, and vehicle energy demand. Therefore, VIPV is best understood as a context-dependent supplementary energy strategy rather than a transformative standalone solution. Its strongest value lies in specific vehicle classes, climates, and usage patterns where on-board generation can reduce charging demand, support operational resilience, or improve distributed self-consumption. The review also proposes minimum reporting requirements for future studies, including annual energy yield, Wh/km contribution, PV area or capacity, mileage assumptions, SOC modeling, and curtailment treatment. The review was not formally registered, and no formal risk-of-bias or certainty assessment was applied. Full article