With the advancement of the Rural Revitalization Strategy and the “Dual Carbon” goals, rural energy systems are exhibiting pronounced multi-energy coupling, a high penetration of renewable energy, and strong load randomness, placing higher demands on the construction of source-load scenarios across multiple time scales. Addressing the limitations of traditional statistical models in generating high-quality short-term source-load scenarios and the tendency of deep learning methods to overlook medium- to long-term seasonal evolution patterns, this paper proposes a hybrid data- and model-driven method for constructing multi-energy source-load scenarios in rural systems. This method establishes a multi-time-scale generation framework encompassing daily scenarios, monthly scenarios, and a full-year 8760-h annual scenario. The proposed approach begins with data preprocessing to form a high-quality sample set. Subsequently, it utilizes a Wasserstein Generative Adversarial Network with Gradient Penalty model (WCGAN-GP) integrated with a temporal attention mechanism to achieve high-fidelity short-term generation of source-load profiles, including photovoltaic, small hydropower, electrical load, and thermal load. Furthermore, it characterizes seasonal evolution features by combining Dynamic Time Warping K-Medoids (DTW–K-Medoids) for typical day extraction and monthly Markov transition modeling. Finally, the annual scenario is continuously spliced and generated through a cross-month boundary cost function and dynamic programming path optimization. Results demonstrate that the short-term scenarios perform excellently across various evaluation metrics, while the monthly and annual scenarios accurately reflect seasonal variations and long-term trends, validating the accuracy and applicability of the proposed method.

Hydraulic fracturing serves as a critical technology for reservoir stimulation in deep coalbed methane (CBM) development, where the mechanical properties of gangue layers exert a significant control on fracture propagation behavior. To address the unclear mechanisms governing fracture penetration across coal-gangue interfaces, this study employs the Continuum-Discontinuum Element Method (CDEM) to simulate and analyze the vertical propagation of hydraulic fractures initiating within coal seams, based on geomechanical parameters derived from the deep Benxi Formation coal seams in the southeastern Ordos Basin. The investigation systematically examines the influence of geological and operational parameters on cross-interfacial fracture growth. Results demonstrate that vertical stress difference, elastic modulus contrast between coal and gangue layers, interfacial stress differential, and interfacial cohesion at coal-gangue interfaces are critical factors governing hydraulic fracture penetration through these interfaces. High vertical stress differences (>3 MPa) inhibit interfacial dilation, promoting predominant cross-layer fracture propagation. Reduced interfacial stress contrasts and enhanced interfacial cohesion facilitate fracture penetration across interfaces. Furthermore, smaller elastic modulus contrasts between coal and gangue correlate with increased interfacial aperture. Finally, lower injection rates effectively suppress vertical fracture propagation in deep coal reservoirs. This study elucidates the characteristics and mechanisms governing cross-layer fracture propagation in coal–rock composites with interbedded partings, and delineates the dynamic evolution laws and dominant controlling factors involved. The findings provide critical theoretical insights for the optimization of fracture design and the efficient development of deep coalbed methane reservoirs.

The deep coal reservoir in Linxing-Shenfu block of Ordos Basin is an important part of China’s coalbed methane resources. In the process of reservoir reconstruction, the artificial fracture morphology of coal seam with gangue interaction is significantly different, which affects the efficient development of coalbed methane resources in this area. In this paper, the surface outcrop of Linxing-Shenfu block is selected, and three kinds of interaction modes between gangue and coal seam are set up, including single-component coal rock sample, coal rock sample with different thicknesses of gangue layer and coal rock sample with different numbers of gangue. Through true triaxial physical simulation and three-dimensional discrete element numerical simulation, the law of artificial fracture initiation and propagation in multi-gangue interaction coal seam is analyzed in depth, and the hydraulic fracture initiation and propagation mode under different interaction modes of gangue layer in Linxing-Shenfu deep coal reservoir was clarified. The research shows that the initiation of artificial fractures in a single coal seam is affected by geological-engineering factors. The maximum principal stress dominates the direction of fracture propagation, and the stress difference controls the fracture morphology. When the stress difference is 2 MPa, the fracture morphology is complex, which is easy to connect to the weak surface of coal and rock cleat, and the fracture morphology of the stress difference is mainly a single main fracture. After the thickness of the gangue layer is increased from 2 to 5 cm, it is difficult for the artificial fracture to penetrate the layer vertically after the fracture initiation, and the effective transformation area of the reservoir is limited. The more the number of gangue layers, the greater the hydraulic energy consumption in the process of fracture propagation, and the more difficult the fracture propagation.

Micro Gas-Cooled Reactor (MGCR) has garnered attention in relevant domains, owing to its advantages of miniaturization and transportability, which is capable of providing stable electrical power to off-grid and special regions. As a typical multi-physics coupled system, a dynamic model for the MGCR integrating nuclear, thermal-hydraulic, mechanical, and electrical subsystems was developed in this study using the multi-physics modeling language Modelica. Steady-state validation results indicate that the maximum deviation between the simulated values and the design parameters is merely 1.05%. Meanwhile, transient validation demonstrates a high degree of consistency with the outcomes generated by the MGT-3D code, with the relative error of the maximum temperature remaining below 3%. Based on the established dynamic model, transient analyses were carried out for three typical accident scenarios: reactivity insertion accidents, loss-of-flow accidents, and load rejection events. The results demonstrate that in the case of reactivity insertion and loss-of-flow accidents, the reactor can achieve a new steady state or autonomously reduce power through its inherent feedback mechanisms, thereby ensuring operational safety; specifically, the maximum temperature rise at the reactor outlet reaches 33.89°C during a reactivity insertion accident and 23°C in the event of a loss-of-flow accident. For load rejection scenarios, the opening of the isolation valve can effectively protect the helium turbine: upon a complete loss of external load, the rotor speed increases by up to 5.6%, which does not exceed the safety limits. This study verifies the inherent safety of MGCRs under accident conditions and demonstrates the application potential of Modelica in multi-physics dynamic analyses of such advanced nuclear systems.

The rapid digitalization of the energy sector has led to the deployment of large-scale smart metering systems that generate high-frequency time series data, creating new opportunities and challenges for energy anomaly detection. Accurate identification of anomalous patterns in building energy consumption is essential for optimizing operations, improving energy efficiency, and supporting grid reliability. This study investigates advanced feature engineering and machine learning modeling techniques for large-scale time series anomaly detection in building energy systems. Expanding upon previous benchmark frameworks, we introduce additional features such as oil price indices and solar cycle indicators, including sunset and sunrise times, to enhance the contextual understanding of consumption patterns. Our comparative modeling approach encompasses an extensive suite of algorithms, including KNeighborsUnif, KNeighborsDist, LightGBMXT, LightGBM, RandomForestMSE, CatBoost, ExtraTreesMSE, NeuralNetFastAI, XGBoost, NeuralNetTorch, and LightGBMLarge. Data preprocessing includes rigorous handling of missing values and normalization, while feature engineering focuses on temporal, environmental, and value-change attributes. The models are evaluated on a comprehensive dataset of smart meter readings, with performance assessed using metrics such as the Area Under the Receiver Operating Characteristic Curve (AUC-ROC). The results demonstrate that the integration of diverse exogenous variables and a hybrid ensemble of traditional tree-based and neural network models can significantly improve anomaly detection performance. This work provides new insights into the design of robust, scalable, and generalizable frameworks for energy anomaly detection in complex, real-world settings.

This paper proposes a cooperative fault recovery method for distribution networks and microgrids considering dynamic network reconfiguration. First, an optimization model is constructed with the objectives of minimizing electricity purchase costs, voltage deviations, network losses, load shedding, and microgrid operation costs. The model thoroughly considers operational conditions, including power flow constraints, security operation constraints, and network reconfiguration constraints, within the distribution network. Second, to address the privacy protection requirements of distributed energy resources (DERs) in microgrids, a k-order approximation method is employed to establish an operation feasible region (OFR) model for microgrid operational flexibility resources. This model characterizes the operational characteristics of microgrids through power and energy boundary constraints. Finally, to solve the cooperative optimization problem between the distribution network and microgrids, a non-iterative solution method based on equivalent projection is proposed. This method decomposes the original problem into a distribution network-level optimization and microgrid-level subproblems, eliminating internal variables through equivalent projection technology to achieve cooperative optimization under privacy protection. Simulation results demonstrate that the proposed method can effectively achieve cooperative optimization of distribution network fault recovery and microgrid scheduling while protecting microgrid privacy information, significantly improving the economy and reliability of distribution network fault recovery.

The intermittency of wind power poses severe challenges to the safe and stable operation of power grids, while conventional forecasting models are deficient in prediction accuracy and adaptability to variable weather conditions. To address these issues, this study proposes an adaptive short-term wind power forecasting model integrating affinity propagation (AP) clustering and a black-winged kite algorithm (BKA)-optimized temporal convolutional network-bidirectional long short-term memory (TCN-BiLSTM) hybrid architecture. First, mutual information was employed to screen key meteorological features, and AP clustering categorized historical data into six distinct weather scenarios. A scenario-specific TCN-BiLSTM model was then constructed for each cluster: TCN was utilized to capture multi-scale local temporal features, while BiLSTM modeled global sequence dependencies, with BKA implementing global hyperparameter optimization. Final predictions were generated by invoking the model corresponding to the nearest cluster center. Comparative experiments against baseline models (LSTM, BiLSTM, CNN-LSTM, TCN-BiLSTM) demonstrate that the proposed model achieves remarkable performance gains: normalized root mean square error (nRMSE), normalized mean absolute error (nMAE), and normalized mean square error (nMSE) are reduced by over 1.27%, 1.73%, and 12.09%, respectively, with the coefficient of determination (R2) reaching 0.9084. This verifies substantial improvements in prediction accuracy and data fitting capability. The scenario-based modeling framework combined with intelligent hyperparameter optimization effectively enhances the model’s adaptability to complex weather, confirming its high accuracy and strong generalization for wind power forecasting tasks.

Building sectors in China contributes more than 40% of the total national energy consumption, and thus efficient control of energy is important in meeting the 2-carbon target. This paper proposes QEMES–ASHP (Quantum-Enhanced Multi-Physics Energy Simulation with Air Source Heat Pump integration) as a real-time platform, combining thermal, electrical, and fluid dynamics modelling with adaptive control using AI. The main innovations of the framework include: (i) real-time multi-physics coupling, in which modelling heat–power–airflow interactions under dynamic climate conditions, (ii) machine learned system switching in solar–ASHP–PCM systems, and (iii) fractal-based time modelling, which uses a Brownian motion at fractional scales to account for energy demand fluctuations between microsecond and season scales and quantify uncertainty in a Bayesian uncertainty quantification for robust predictions. The quantum enhancement is the utilization of a variational quantum linear solver (VQLS), accelerating large-scale coupled matrices, with up to 20% faster convergence than classical solvers, and is compatible with classical HPC resources. Integration-first simulation paradigm is a name that characterizes the steady integration of various physics, predictive AI control, and uncertainty quantification into an integrated, real-time operation structure. Validation was carried out at a 200 m2 IoT-instrumented test structure in 5 Chinese climate zones (hot-humid, hot-dry, temperate, cold, severe cold) and a 150+ sensor count constantly recording data over twelve months. Based on QEMES–ASHP as the reference, the findings indicate 31%–35% yearly energy savings, 15%–30% peak load saving, and PMV-based thermal comfort with the range of ±0.15, which is better than an advanced façade, radiant cooling, and thermal system with TES (p < 0.001). Computational speed was 20 percent faster, with the economic analysis paying back in 4.2 years. Life-cycle analysis suggests a 30%–40% decrease in the number of CO2 emissions, which will be directly part of the carbon neutrality in China in 2060. QEMES–ASHP illustrates a simulation methodology that is both holistic and replicable, providing practical recommendations to architects, HVAC engineers, and policymakers who design next-generation net-zero smart buildings in a variety of Chinese climates.

Aiming at the contradiction between green energy consumption and diesel dependence in temporary construction camps under the condition of “weak data-weak communication”, this paper puts forward a collaborative framework of off-grid light storage and firewood storage with tight coupling of “prediction-scheduling”. The prediction layer constructs a lightweight IOOA-CNN-BiLSTM-Markov model: CNN extracts the spatial characteristics of tower crane shadow and cloud cluster, BiLSTM captures the bidirectional time series dependence, Markov residual compensates the non-stationary disturbance, and uses the improved Osprey algorithm to complete the small sample superparameter self-tuning at the edge of ARM, thus realizing the error-oriented compression in the scene of lack of weather channels. The scheduling layer is designed to improve the affine adjustable rolling optimization driven by the snake optimization ISO. With the prediction interval-scene dual mode as the input, the start-stop frequency of diesel and the energy storage SOC are modeled as integer-continuous variables synchronously, and the three-objective MILP of “minimum diesel + minimum load loss + minimum light rejection” is solved in seconds, thus achieving the multi-objective coordination of green power priority, energy storage arbitrage and diesel compensation. By building a photovoltaic, energy storage and diesel hardware-in-the-loop platform and loading a typical temporary site with an impact load of 80–400 kW, the system verifies the adaptability and deployment of the proposed framework under the extreme conditions of “zero power grid, zero history and weak sensing”.