Enhancing the coal adaptability of plasma ignition devices holds significant importance for assisting thermal power plants in energy conservation, emission reduction, and achieving carbon peaking and carbon neutrality goals. Currently, graded plasma combustors encounter operational challenges including combustion instability and flameout when burning low-volatility lean coal in power plants. To address these issues, this study employed numerical simulation methods. Initially, a structured mesh of the graded plasma combustor was established and its reliability was verified. Subsequently, using the controlled variable approach, the effects of three critical operating parameters - plasma power, primary air velocity, and pulverized coal concentration - on the lean coal combustion process in graded plasma combustors were systematically investigated. Finally, the operating parameters of the plasma burner were optimized to resolve ignition and combustion challenges associated with lean coal. The research results demonstrate that plasma power is the critical factor influencing lean coal ignition. To achieve stable ignition and combustion of lean coal in the plasma burner, the plasma power should not be lower than 150 kW. The primary air velocity affects the combustion temperature during the initial ignition stage, with an optimal range identified between 22-25 m/s. Furthermore, pulverized coal concentration proves to be a significant factor for successful ignition, where maintaining a relatively high concentration is essential for stable combustion - the coal concentration should not be less than 0.3 kg/kg. These findings provide important operational guidance for power plants utilizing plasma burners in lean coal ignition applications.

Traditional home energy management methods mainly focus on the energy management within individual households, overlooking considerations of comfort and the interaction with the grid. To enhance the capabilities of home energy management, a two-layered home energy management strategy is proposed. The upper layer considers the operational constraints and costs of the distribution network and the lower layer aims to fulfill users' electricity consumption preferences and cost requirements, achieving optimal household energy management under the premise of interacting with the grid. Acknowledging the limitations of conventional power flow calculations for distribution networks and indoor thermal response algorithms, data-driven methods are introduced to improve the accuracy in assessing both the grid's state and indoor thermal conditions. Furthermore, a natural aggregation algorithm is employed to increase the precision of the optimization process. Extensive experiments are conducted to evaluate the accuracy of the proposed algorithm, and the results convincingly demonstrate the effectiveness and scientific rigor of the proposed strategy.

The frequency of the new electricity system is facing severe tests such as reduced rotational inertia, enhanced fault impact, and insufficient frequency regulation resources. Distributed resources, with distributed energy storage as a typical example, have become a new means and effective supplement for system frequency regulation. However, the contradiction between the autonomous demands of users and the deterministic requirements of the power grid restricts the application of distributed energy storage in the auxiliary frequency regulation scenario. To address this issue, first, the decision-making behavior psychology of distributed energy users is analyzed through prospect theory, and a more realistic distributed energy participation in frequency regulation simulation environment is constructed based on the user incentive response characteristic curve under non-rational decision-making. Secondly, a flexible control strategy for broadcasting dynamic incentive signals to distributed energy storage users is proposed. The nonlinear problem of frequency regulation is transformed into a system parameter estimation problem after dynamic linearization, and a model-free adaptive incentive control method is adopted to solve the uncertainty of user response and the difficulty of user cluster modeling. The cluster response model is dynamically linearized through real-time data of the cluster control system and the optimal real-time incentive signal is output. Finally, a case study of a single-area power system is conducted in MATLAB/Simulink. Under both step load fluctuation and continuous load fluctuation scenarios, the proposed method has good frequency tracking performance, with the maximum frequency deviation controlled below 0.0015 p.u., thereby verifying the adaptability and effectiveness of the proposed incentive control method.

Interactive traffic scenarios—characterized by high-dimensional complexity and elevated safety stakes—constitute a foremost validation bottleneck for autonomous vehicles (AVs). Recent learning-based scenario-generation paradigms—encompassing data-driven synthesis, adversarial generation, knowledge-guided generation, and large language models (LLMs)—have demonstrated superior fidelity and coverage, substantially improving both the quality of interactive-scenario corpora and the efficiency of safety-critical testing. To systematically chart the advances, core techniques, and remaining impediments in this domain, this paper presents a comprehensive survey of learning-oriented approaches to generating interactive scenarios for AVs, delineating a coherent technical roadmap and future research directions.First, we dissect the fundamental attributes of interactive scenarios and benchmark prevailing open datasets, elucidating their respective capacities to underpin learning-based generators. Concurrently, we review pertinent standards, formal scenario-description languages, and evaluation metrics, analyzing their expressive power over interactive behaviors. Second, we taxonomize existing methods into two strands—conventional learning frameworks and LLM-based paradigms—synthesizing their algorithmic pipelines and empirical generation performance. Finally, we distill cross-cutting challenges residing in data curation and methodological design, and outline prospective research avenues.Our analysis reveals that traditional learning methods struggle to simultaneously guarantee realism and diversity, and remain inefficient in discovering and synthesizing safety-critical edge cases. LLM-based approaches, while exhibiting remarkable proficiency in semantic comprehension and complex logical composition, suffer from sluggish inference, hallucination, misalignment with physical plausibility, and opaque decision-making that undermines interpretability. Moreover, the field at large is constrained by datasets that lack fine-grained interaction-critical annotations and by the absence of unified severity metrics for scenario endangerment. We contend that, despite preliminary successes in interactive modeling, substantial enhancements in generation fidelity and generalizability are imperative to meet the stringent demands of real-world AV deployment.

This paper proposes a Transformer-based traffic prediction model named ST-Trans and employs ST-Trans to develop a predictive motion planner for intelligent connected vehicles. ST-Trans utilizes Transformer to extract spatial-temporal evolution patterns from real-time vehicle data and lane segment structural information provided by dynamic high-definition maps, predicting the traffic state of lane segments. Meanwhile, ST-Trans further enhances prediction accuracy by incorporating connectivity among lane segments and phase information at intersections. The prediction result demonstrates that ST-Trans outperforms the best baseline model by 12.2%, 12.1%, and 3.55% in mean absolute error, root mean square error, and accuracy, respectively. Based on the prediction results of ST-Trans, the predictive motion planner significantly reduces the computational complexity of motion planning tasks so that it swiftly computes the path plan and speed curve among road sections ahead. This study combines SUMO and CARLA to validate the predictive motion planner and the simulation results demonstrate that the ST-Trans-based predictive motion planner is capable of achieving predictive path and speed planning, and it surpasses traditional motion planners in terms of safety, efficiency, comfort, and computational speed.

To address the issue of decreased recognition accuracy caused by false and missed detections of small target traffic signs, a small traffic sign recognition algorithm based on CGT-YOLO is proposed. First, the Context Enhancement Module is introduced to replace the SPPF module in the YOLOv5s network, enhancing the feature representation of small traffic signs through convolutions with varying dilation rates. Next, the Global Attention Module is added after the Concat module in the YOLOv5s backbone network. The GAM extracts the CAM-enhanced features and further strengthens the global interaction between channels and spatial dimensions using 3D permutation, multi-layer perceptron, and convolutional spatial attention. This process highlights the features of small target traffic signs and mitigates the negative effects of complex backgrounds and long distances. Finally, the TSC decoupling head is utilized to separate the features for classification and localization tasks. This module generates semantically rich, low-resolution classification feature maps and high-resolution localization maps containing boundary information, effectively resolving the feature conflicts between the two tasks. Experimental results show that the improved model has improved in all indicators: the miss rate and false positive rate decreased by 12.1% and 11.6%, respectively, and the mAP(0.5:0.95) increased by 2.6%. These improvements effectively reduce false positives and false negatives, significantly, enhancing the recognition accuracy of small traffic signs.

Under the global low-carbon development trend, battery electric buses (BEBs) have emerged as a crucial component of urban transportation systems due to their environmental benefits. However, their large-scale deployment faces challenges posed by limited driving ranges, demanding coordinated planning of charging infrastructure and operational schedules. Existing studies predominantly treat charging station siting and timetabling as separate optimization problems, overlooking their inherent interdependencies. Furthermore, current approaches focus primarily on single-depot configurations or small-scale networks, resulting in limited generalizability. To address these gaps, this study proposes a spatio-temporal network framework for multi-depot BEB systems, establishing a joint optimization model that integrates charging station siting and timetabling. The model aims to minimize the total system cost while incorporating constraints such as charging station construction, trip sequence continuity, battery charge maintenance, vehicle scheduling, and charging pile-vehicle matching. An enhanced memetic algorithm combining improved genetic operators and local search strategies is developed to solve the proposed model. Case studies on a real-world bus network in Chancheng District, Foshan City, validate the effectiveness of the framework across different problem scales. Results demonstrate that: (1) The proposed algorithm achieves significant total cost reductions compared to traditional genetic algorithms and simulated annealing methods in both small- and large-scale scenarios; (2) Charging station planning indirectly influences the total system cost by affecting timetabling efficiency, thereby justifying the necessity of joint optimization; (3) The framework enriches the theoretical foundation of BEB system optimization and provides practical decision-making references for both transit planners and operators.

An integrated "battery-vehicle matching" strategy was proposed to address key issues for electric buses in cold regions. These issues include accelerated battery degradation and significant variability in driving range. In this strategy, vehicles are equipped with different capacity batteries based on seasonal temperatures and route demands. This method allows for more optimized vehicle scheduling. A mixed-integer programming model was developed to minimize total operator costs. The model calculated battery acquisition costs, daily maintenance, and charging fees that considered time-of-use rates. A comprehensive function, which quantified both cyclic and calendar battery degradation, was integrated into the cost calculations. This complex model was solved using a specially designed Hybrid Genetic Algorithm (HGA). A study was conducted centered on the actual operating bus routes in Harbin. The results showed significant improvements. The total annualized cost was reduced by 17.8%, and the required fleet size was decreased by 16.4%. Consequently, vehicle acquisition and maintenance costs were saved by 11.5% and 16.8%, respectively. It was also noted that a dual-battery configuration cut annual battery degradation costs by 39.7%, effectively extending the service life of the batteries. Finally, a sensitivity analysis revealed that battery aging can be further slowed by relaxing the maximum state of charge during mild seasons. Additionally, the applicability of the model in larger-scale scenarios is discussed in the concluding section of the instance analysis. This study provides a valuable model and tool for operators to develop seasonal strategies for sustainable operations in cold climates.

In view of the insufficient carrying capacity of highway distribution grids and the mismatch between re-newable energy generation and consumption loads, the use of mobile energy storage vehicles for energy interexchange between service areas can promote the consumption of renewable energy. In order to improve the real-time responsiveness of mobile energy storage, an optimization strategy is proposed that considering the traffic status under the constraint of the average loss rate of electric vehicle battery swapping service. A Markov chain model is developed to characterize the traffic status between service areas and the energy states of battery swapping station. By analyzing the average transportation cost of mobile energy storage and the average loss rate of battery swapping service, an optimization problem for mobile energy storage vehicle dispatching is formulated and solved to obtain the optimal scheduling strategy and its parameters. The simulation results validate the proposed strategy has a dual-threshold structure with traffic status and energy state. Compared to the greedy strategy and Q-learning algorithm, the average transportation cost of mobile energy storage is reduced by 17.23% and 8.89%, respectively.

Investigating the off-campus travel mode choice behavior of college students is crucial for enhancing the off-campus travel environment of college students, increasing their social participation, and further promoting their physical and mental health development. However, previous studies neglect the influence of peer groups on individual travel mode choice and lack a unified standard for identifying effective peer groups among college students. To explore the impact of peer effect on off-campus travel mode choice behavior of college students, this study defined effective groups using a roommate relationship quality scale and collected college students’ off-campus travel preference data under different travel scenarios based on questionnaire survey, identifying effective groups from a dorm social network perspective. Multiple network economics linear models were constructed by introducing a peer matrix adjusted for relationship strength, grounded in relationship quality and psychological traits, potential endogeneity issues were addressed though combining fixed effect and two-stage least squares, and the best model to identify and measure the peer effect was determined through model evaluation. Based on the model calibration results, the influence and potential mechanism of dormitory groups on individual off-campus travel mode choice of college students were analyzed. The results show that ① the local average model that incorporates peer relationship strength has better robustness in identifying and measuring peer effect compared with generalized econometric model and local aggregate model. ② With the increase of travel distance and the relaxation of travel time constraints, the influence of endogenous peer effect on individual travel mode choice behavior decreases. ③ The factors such as monthly living expenses, number of family members, driver's license, openness to experience and environmental awareness related to peers have significant exogenous peer effect on individual travel mode choice for college students. ④ The endogenous peer effect of conformity is more pronounced for extrovert individuals compared to introvert individuals. In leisure travel scenarios, the endogenous peer effect of conformity is also significantly stronger for individuals scoring high on openness to experience relative to their conservative counterparts. The findings expand the theoretical boundaries of peer influence in travel behavior and provide empirical evidence for refined guidance and subgroup-targeted interventions to promote low-carbon travel among college students.

To explore the impact of crew integrated optimization on the cost of crew scheduling and efficiency of crew utilization in cross-line operation, a spatio-temporal network was constructed to represent the rules for crew segment connections and shift paths. To determine the rules of crew segment connections, a network model was constructed with the optimization objectives of minimizing the number of shifts and idle time. The network graph search and improved column generation algorithm were designed to obtain shift paths. A mixed crew rostering approach was proposed to verify the impact on crew scheduling. The case study was carried out based on the crew data of some metro lines in a certain city. The results show that the integrated crew scheduling is able to increase the average work efficiency by 1.5%~2.3%. The deadhead segments and crews decrease by 12.18%~24.45% and 1.16％~11.46％. Compared with the phased optimization, the integrated optimization of crew scheduling improves the average work efficiency of shifts and utilization efficiency of crews. It reduces the number of daily shifts and average number of deadhead segments. The integrated optimization approach with the mixed crew rostering mode is able to adapt to the flexible shift cycles and spatiotemporal distribution differences of crew segments, which is conducive to ensuring the balance of shifts and efficiency of crew utilization.

As an advanced non-destructive three-dimensional (3D) imaging detection technique, X-ray Computed Tomography (CT) enables the visualization of internal structures within samples. It operates based on the interaction mechanisms between X-rays and matter, integrated with sophisticated computed tomography principles. Through detectors, it captures signals transmitted through the sample, which are subsequently processed via algorithms to reconstruct tomographic images for imaging purposes. Endowed with advantages such as high-density resolution and facile digital processing, this technology has achieved significant breakthroughs in domains including medical diagnosis and industrial inspection.  In the field of materials science, the value of X-ray CT technology is particularly pronounced: it not only facilitates in-situ 3D quantitative analysis of internal defects (e.g., pores, cracks) in structural materials but also dynamically tracks the damage evolution processes of materials under complex environments such as loading and corrosion. Via multi-scale (ranging from nanoscale to centimeter-scale) and multi-modal (including morphology, composition, and orientation) collaborative characterization, it can effectively unravel the structure-activity relationships of new energy materials, thereby providing crucial foundations for catalyst design and battery optimization.  This paper systematically synthesizes the core principles of X-ray CT technology, focuses on its frontier applications in structural materials and new energy materials, analyzes technical strengths and existing bottlenecks with specific case studies, and envisions future breakthrough pathways. These explorations not only offer directions for technological innovation to researchers but also contribute to enhancing the independent research and development capabilities of China's high-end detection equipment, holding significant strategic importance for strengthening core competitiveness.

With the increasing severity of environmental noise pollution, concrete noise barriers have been widely used in noise control due to their durability and cost-effectiveness. However, conventional concrete barriers have limitations in sound absorption and insulation, making it difficult to simultaneously achieve lightweight, multifunctional, and sustainable performance. Recent studies have incorporated foaming agents and porous lightweight aggregates into cement-based systems, which improves pore structures and sound energy dissipation, thereby enhancing the noise reduction capacity of concrete at the material level. 3D printing has offered new opportunities for constructing noise barriers by enabling complex geometries, lightweight structures, and structural customization, which promote sound scattering and energy dissipation, further improving both sound absorption and insulation performance. This article summarizes the key factors and regulation strategies influencing the acoustic performance of concrete, analyzes the role of 3D printing in tailoring pore architecture, interfacial bonding, and surface textures, and reviews representative engineering applications. The analysis indicates that 3D-printed noise-reducing concrete shows considerable potential for roadway noise barriers and building acoustics, although challenges remain in material printability, interlayer bonding, and pore structure stability. Future research should focus on the application of low-carbon and sustainable materials, multi-scale structural design, and durability assessment to enable the high-performance and practical applications of 3D-printed noise-reducing concrete.

In order to realize the resource utilization of solid waste and improve the performance of recycled concrete blocks in severe cold regions of Qinghai-Tibet. On the basis of 100% recycled coarse aggregate, the effects of recycled fine aggregates and active admixtures on the impermeability, water resistance and frost resistance of high-volume solid waste recycled concrete load-bearing blocks were studied. The microstructure was analyzed by SEM and NMR. The results show that the permeability and water absorption of recycled concrete load-bearing blocks gradually increase with the increase of recycled fine aggregate, and decrease first and then increase with the decrease of fly ash and slag mixed ratio, while the frost resistance and softening coefficient change oppositely. When the replacement rate of recycled coarse aggregate is 100 % and the mix ratio of fly ash and slag is 3 : 1, the compressive strength is 11.77 MPa, the flexural strength is 3.89 MPa, the softening coefficient is 0.99, the water absorption rate is 0.7 %, the mass loss rate after 50 freeze-thaw cycles is 2.2%, and the loss rates of compressive strength and flexural strength are 10.2% and 13.9% respectively, which meet the requirements of load-bearing and durability in severe cold regions and can be used as a recommended mix ratio. Microscopically, alkali excitation promotes the secondary hydration of the composite cementitious materials, generates more hydration products to optimize internal pores, and the concrete structure is dense. With the increase of the number of freeze-thaw cycles, the number of pores in the block gradually increases, and the micropores and mesopores gradually derive into large pores and fissures, and the performance of the block deteriorates. Based on the performance-cost-carbon emission analysis, 100 % recycled coarse aggregate is not conducive to low-carbon emission reduction. The recommended mix ratio of mixed recycled aggregates and active admixtures has a carbon emission reduction rate of 31.03%.

The abundant hydrophilic hydroxyl groups present on paper fibers surfaces limit its applicability in barrier packaging. Currently, most petroleum-based materials used to enhance the barrier properties of paper are poorly degradable. In contrast, alkali lignin exhibits inherent biodegradability, hydrophobicity, and flame retardancy, offering distinct advantages for the development of sustainable barrier packaging materials. To investigate the influence of alkali lignin on the hydrophobic performance of packaging paper, alkali lignin (A-Lig) was used as the raw material in this study. Esterification reactions were carried out using palmitoyl chloride and stearoyl chloride, yeilding lignin palmitate (Lig-P) and lignin stearate (Lig-S), respectively. Subsequently, coating solutions were formulated based on the three types of lignin (A-Lig, Lig-P, and Lig-S) and sprayed onto base paper to fabricate superhydrophobic paper. The chemical structures, micromorphologies, and thermal properties of A-Lig, Lig-P, and Lig-S were systematically characterized using a suite of analytical techniques, including Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), nuclear magnetic resonance (NMR), scanning electron microscopy (SEM), and thermogravimetric analysis (TGA). Meanwhile, the microstructure, hydrophobic properties, and mechanical strength of both the base paper and coated papers comprehensively evaluated through static contact angle, rolling angle, water absorption, surface water stability, self-cleaning property, and mechanical performance measurements. The results showed that the hydroxyl groups in the esterified lignin were effectively substituted, with aliphatic chains successfully grafted onto the lignin backbone. Additionally, the relative content of C-O bonds decreased, while the thermal stability of the modified lignin was reduced. The static contact angles of the base paper (Bas-P), A-Lig-coated paper (Lig-P1), Lig-P-coated paper (Lig-P2), and Lig-S-coated paper (Lig-P3) were measured as 45.9°, 83.5°, 150.8°, and 151.6°, respectively. Notably, the rolling angles of Lig-P2 and Lig-P3 were 9.3° and 3.5°, respectively, satisfying the established criteria for superhydrophobicity. Moreover, a petal-like micro-nano hierarchical roughness was observed on the surfaces of Lig-P2 and Lig-P3. Compared with Bas-P, Lig-P2 and Lig-P3 exhibited a slight decrease in tensile strength but a pronounced increase in elongation at break, indicating improved flexibility of the paper.