ABSTRACT Rockbursts are typical dynamic disasters in underground coal mines. Their formation and occurrence are influenced by multiple factors, including geological conditions, mining disturbances, and microseismic activities. Machine learning models can support early warning by forecasting high‐energy microseismic (MS) activity associated with elevated rockburst risk. However, many of these models are black boxes that are difficult for humans to understand, which may diminish their credibility and constrain their practical applicability. To establish interpretable models for short‐term rockburst prediction considering multiple factors, we assess five tree‐based models (decision tree, random forest, LightGBM, XGBoost, and CatBoost) using field databases from Gao Jiapu coal mine and utilize the model explanation method SHAP to interpret the models. A day is labeled as rockburst risk if its daily maximum MS energy exceeds 1 × 10⁵ J, otherwise no risk. The results revealed that CatBoost exhibited the highest overall performance, followed by LightGBM, and both the random tree and XGBoost performed better than the decision tree. When considering the influence of time window length, all models displayed optimal performance under a 3‐day time window. Specifically, CatBoost achieved F1 scores of 0.856, 0.778, and 0.807 for the 3‐day, 5‐day, and 10‐day time windows, respectively. Global feature importance analysis showed that the average energy, energy deviation, and cumulative energy were the three most important predictors. Furthermore, local SHAP values were drawn to reveal the complex underlying relationships between the 12 factors and model predictions. The results show that factors such as energy‐related factors, coal depth, and working face advance positively contribute to rockburst risk prediction. Conversely, fold distance, total frequency, and source concentration degree negatively impact the model outcomes. The findings suggest that incorporating CatBoost and the SHAP method holds great potential in developing interpretable models for accurate rockburst risk prediction.

ABSTRACT Energy poverty remains a major challenge of the 21st century. It reflects a trade‐off between satisfying current energy demands and protecting the environment. Similarly, in North Gondar, the sector continues to challenge socioeconomic development and adversely affects household health and the environment, with limited progress in switching to clean energy sources. Household fuel preference plays a central role in shaping this transition. This study examined the extent of solid biofuel usage and household energy choice behavior in Northwest Ethiopia, using data collected from 384 randomly selected households and a multivariate probit model to capture the interrelated nature of fuel choices. The results indicate a heavy reliance on solid biomass, which accounts for 90% of cooking energy use and 76% of lighting energy consumption. Empirical findings reveal that households living in urban areas, with higher incomes and better educational backgrounds, are more likely to adopt cleaner energy sources. Older households, those living far from markets, and households with larger livestock holdings continue to rely on traditional biomass fuels. The study concludes that household energy choice in the area follows a fuel‐stacking pattern rather than a simple linear transition along the energy ladder. On the basis of the insights, the study recommends strengthening public awareness campaigns, prioritizing pro‐poor energy transition measures, supporting afforestation efforts, and promoting environmentally friendly biomass practices to improve household energy use and accelerate the transition to clean energy sources in Northwest Ethiopia.

ABSTRACT Tight sandstone gas (TSG), an unconventional oil–gas resource, has heterogeneous reservoirs. Traditional classification and evaluation methods fail to fully reveal reservoir characteristics and complex reservoir‐productivity links. As a key source of tight sandstone, the Taiyuan Formation in the Ordos Basin exhibits significant variations in reservoir quality due to its marine–terrestrial transitional sedimentary setting, necessitating the establishment of an effective classification and evaluation system. This study focuses on Block A's Taiyuan Formation tight sandstone reservoirs in the Ordos Basin, using core observations, thin‐section petrography, scanning electron microscopy, and high‐pressure mercury intrusion experiments to build a machine learning (ML) model for reservoir evaluation and improve the accuracy of reservoir evaluation. Results demonstrate that the K ‐means algorithm has the optimal classification results based on a variety of internal evaluation metrics. The tight sandstone reservoirs in the study area are classified into three distinct types: Type I reservoirs are mainly lithic quartz sandstone/lithic feldspathic sandstone, showing significant compaction, with large pore throats and low displacement pressure. Type II, mostly feldspathic lithic sandstone, acts as a transitional reservoir. Type III, predominantly lithic feldspathic sandstone, is most affected by cementation, featuring fine‐grained pore throats and high displacement pressure. Furthermore, evaluation results based on confusion matrices demonstrate that the Bayesian‐optimized Random Forest model achieves a classification accuracy of 92.76% in log interpretation, significantly outperforming other models. Through gray relational analysis, the nonlinear correlations between productivity and reservoir parameters are clarified. The prediction results demonstrate over 80% spatial consistency with actual gas reservoir distributions, effectively reflecting the production capacity of the tight sandstone reservoirs in the study area. The ML‐driven evaluation approach proposed in this study not only provides robust technical support for the exploration and development of TSG reservoirs and effectively delineates the distribution of favorable zones, but also offers practical guidance for the efficient exploration of hydrocarbon resources in geologically analogous settings.

ABSTRACT In contemporary society, improving energy transition efficiency (ETE) has become a necessary measure to ensure the normal and sustainable functioning of society. Aiming to inform region‐specific “dual‐carbon” governance, this study quantifies the drivers of energy‐transition efficiency across 30 Chinese provinces from 2013 to 2022. Applying Super‐SBM, GML decomposition and a spatial Durbin model, we test how resource endowment and digital transformation affect ETE and whether spatial spill‐overs call for differentiated policies. The results are: (1) National ETE exhibits an “east‐high, west‐low” pattern yet inter‐regional gaps narrowed significantly after 2019, with technological progress contributing most; (2) resource abundance raises local ETE and exerts an even stronger positive indirect effect on neighboring provinces, indicating an “endowment advantage” rather than a curse; (3) digitalization enhances local efficiency but generates negative spill‐overs, suggesting a competitive “digital siphoning”; (4) regional heterogeneity is pronounced: eastern digital benefits spill over weakly while resource effects are modest, central provinces gain most from resource‐driven diffusion, and western regions experience strong spill‐overs from both factors. These findings affirm that one‐size‐fits‐all energy policies are inadequate; instead, tailored strategies that leverage local resource strengths, coordinate digital investment, and manage cross‐regional externalities are essential for accelerating an equitable and efficient national energy transition.

ABSTRACT In the fully mechanized caving mining of deep coal seams, the dam structure has the characteristics of small size, large burial depth, and significant height‐to‐depth ratio. Traditional methods are difficult to accurately calculate its movement and deformation. Therefore, this study proposes to separate the analysis of the surface and the dam structure's movement and deformation, quantitatively invert the dam deformation state through the surface deformation shape, obtain the deformation law of the dam under different advance lengths, and establish a spatio‐temporal inversion mechanism of local amplification effect. Taking the impact detection of a certain mine dam working face as an example, the study shows that the dam has undergone an alternating process of tensile failure to compressive reduction, but the sequence of tensile and compressive conversion at different parts is different. At the initial stage of the impact, the dam's foot on the anticline side first experiences tensile failure, then affects the upstream slope side, and finally spreads to the middle of the dam; the tensile and compressive conversion also follows the sequence from the anticline side to the upstream slope side and finally to the middle of the dam. Through the local amplification analysis method of replacing coal seam mining with shallow layer mining, the tensile and compressive alternating failure characteristics of the dam evolving from the backwater slope to the water‐facing slope and then to the middle are revealed. The measured maximum settlement value is 7.077 m, with an error of only 0.077 m compared to the simulation result of 7 m for similar materials, verifying the feasibility, rationality, and scientificity of this method. This paper constructs a numerical model that accurately reproduces the deformation of the surface and the dam, analyzes the formation mechanism of the subsidence basin, and determines the excavation volume corresponding to each advance step based on the probability integral method or full‐scale numerical simulation.

ABSTRACT Technological advancements, environmental concerns, and improved electricity management drive the transition to smart grids. With the increasing integration of distributed energy resources (DERs), efficient management is crucial for optimizing both renewable and non‐renewable energy sources. This paper presents a bi‐level stochastic optimization framework to enhance virtual power plant (VPP) operations and improve grid flexibility. At the lower level, VPPs maximize profits by optimizing resource scheduling, managing demand response (DR) programs, and determining power exchanges with the distribution system operator (DSO). The upper level minimizes DSO costs while ensuring network security and reliability. A key feature is the provision of upward and downward flexibility by VPPs at both the individual resource level and in day‐ahead scheduling, enabling the DSO to enhance grid stability and renewable integration. The model, formulated as a stochastic mixed‐integer linear programming (MILP) problem, effectively addresses uncertainties in load demand, renewable generation, and energy prices. Simulation results demonstrate significant improvements in flexibility and economic efficiency, achieving an 89.91% reduction in excess renewable generation through targeted penalty mechanisms. These findings highlight the practical benefits of the proposed framework for VPP operators and DSOs, offering a robust strategy for managing DERs while ensuring stable and secure grid operation.

ABSTRACT Predicting energy consumption helps countries make strategic decisions in many critical areas such as energy management, economic development, energy security, environmental sustainability and infrastructure investments. Therefore, accurate and reliable energy consumption predictions are vital to ensure the sustainability and prosperity of countries. This study aims to contribute to the proper planning of transportation policies and energy management by successfully predicting Türkiye's railway energy consumption. In this direction, energy prediction values were obtained from 18 different machine learning methods using the country's railway line length, number of passengers, freight amount and energy consumption values from 1977 to 2024. To further strengthen the results obtained with these methods, bagging, boosting, stacking and blending ensemble learning methods were utilized. With the improvements, the R‐squared value was increased up to 0.9667 and energy predicting was achieved with very high accuracy. Based on the results obtained from this study, it is possible to provide investment planning more efficiently. In addition, the implementation of energy management strategies, infrastructure planning and sustainable energy policies will be provided more efficiently as a result of obtaining more successful results by using ensemble machine learning methods instead of traditional machine learning methods for energy consumption predictions in different sectors.

ABSTRACT Grass is an abundant and renewable source of cellulose, which makes it a sustainable and cost‐effective source for producing cellulose nanocrystals. Moreover, the extraction of cellulose nanocrystals from grass provides a value‐added product from an otherwise low‐value agricultural waste material, which can contribute to the development of a circular economy. In this study, cellulose nanocrystals (CNCs) were extracted from Hyparrhenia filipendula via citric acid hydrolysis. The Hyparrhenia filipendula stems were pre‐processed through mechanical size reduction and Soxhlet extraction. The extractive‐free stems were fractionated using two solvents: 10 wt% (w/v) sodium hydroxide (NaOH) and deep eutectic solvent of ethylene glycol:citric acid (1:2 molar ratio). The fractionated samples were bleached with acidified sodium chlorite and hydrolyzed with 80 wt% citric acid for 4 h at 120°C in a Parr reactor. The samples obtained at each treatment stage were characterized using standard scientific procedures for chemical composition, morphology, elemental composition, crystallinity, and thermal stability. The results show that CNCs were successfully extracted from Hyparrhenia filipendula via citric acid hydrolysis. The surface morphology of alkali fractionated CNCs was needle‐like, whereas the surface morphology of DES fractionated CNCs was rod‐like. The alkali fractionated and hydrolyzed sample, NaCNC, had the highest cellulose purity (91%), as well as the highest thermal stability. The FTIR analysis proved the removal of non‐cellulosic components in the CNCs, except for the unbleached CNCs that had significant quantities of hemicellulose and lignin. The XRD analysis revealed the presence of characteristic cellulose Iβ in the CNCs, with the UNNaCNC sample (NaOH fractionated, unbleached, acid hydrolyzed sample) having the highest crystallinity index of 81% and the largest crystallite size of 4.20 nm. The properties of the CNCs obtained in this study are comparable to CNCs derived from previously reported lignocellulosic materials. The CNCs from Hyparrhenia filipendula therefore have a wide range of potential applications.

ABSTRACT The HC method for hydropower is a commonly used rock mass quality classification technique in China's hydropower industry. Due to the anisotropic nature of the layered schist in the study area, and the varying angles between different tunnel layers and the tunnel axis, significant discrepancies arise between the HC method's classification results and actual rock mass classifications when these angles are parallel. This study employs uniaxial compression tests on schist to reveal its anisotropic characteristics under loading directions at 0°, 45°, and 90° angles relative to the bedding planes. The compressive strength exhibits a V‐shaped variation with changes in angle between loading direction and schistosity plane, while the elasticity modulus shows a linear decrease as this angle varies. Numerical simulation experiments were conducted to monitor deformations of surrounding rock masses around tunnels. The findings indicate that as the angle between bedding orientation and tunnel axis decreases, both wall and roof deformations increase progressively. Under conditions of 0°, 30°, 45°, 60°, and 90° angles, the ratios of wall deformation values are approximately 1:3.73:4.74:5.44:7.7; whereas for roof deformation values, they are about 1:1.3:1.94:4.7:6.7. When applying traditional HC methods for classifying surrounding rock quality in parallel schist tunnels, a low agreement rate of only 13.33% was observed. However, by incorporating adjustments based on scoring criteria related to major structural plane orientations into numerical simulation results—specifically modifying weights assigned to structural planes—the agreement rate improved significantly to an impressive 100%. These research outcomes effectively enhance both accuracy and applicability in classifying layered rock masses, providing reliable foundations for tunneling construction practices.

ABSTRACT To analyze the oil cavity lubrication characteristics of wind turbines sliding bearings with discontinuous flow in mixed lubrication wear caused by strong sudden changes and heavy loads, this paper aims to improve a method to analyze the pressure and temperature of different oil cavity shapes with different condition to enhance the lubrication reliability. This proposed method combining CFD simulation based on the Finite Volume Method with RNG k – ε model is demonstrated to analyze effectively the oil cavity lubrication characteristics with mixed lubrication wear. The different oil cavity shapes—circular cavity, sector cavity, and cross‐shaped cavity, and different conditions including eccentricity, input pressure, temperature‐viscosity, and surface roughness were analyzed to explore the effects on the maximum pressure, bearing capacity, and temperature. The study accurately evaluates the load‐bearing support performance, temperature rise characteristics with different cavity shapes and different conditions. The results provide valuable references for optimizing lubrication reliability for wind power sliding bearing.