摘要 : Introduction: With the rapid popularization of electric vehicles, their charging load influences the stable operation of the power grid. An accurate prediction of EV charging station load is crucial for optimal resource allocation in power systems. The objective of this study is to address the issue of insufficient accuracy in existing prediction methods, this paper proposes a hybrid prediction model based on Bidirectional Long Short-Term Memory and Adaptive Boosting, aiming to improve the accuracy and stability of medium and long-term EV charging station load forecasting. Methods: The study employs a three-step approach: (1) The pearson correlation analysis was utilized to evaluate multi-dimensional influencing factors and reduce dataset dimensionality' (2) implementation of a BiLSTM neural network for temporal feature extraction and preliminary prediction' and (3) application of the Adaboost algorithm to construct a weighted combination of strong classifiers. The model's effectiveness was validated through comprehensive simulation tests using real-world charging station data. Results: The proposed Pearson feature selection-based BiLSTM-Adaboost model outperforms traditional benchmark models (LSTM and SVM), effectively reduces data redundancy through feature selection, achieves better performance in key indicators (MSE, RMSE, and MAPE), and demonstrates strong generalization ability and robustness while maintaining high accuracy. Discussion: Experimental results demonstrate that the proposed method effectively extracts key features of charging loads, achieving superior prediction accuracy and generalization ability compared to traditional methods. This provides a reliable decision-making tool for power grid operation, effectively supporting the resilience planning needs of urban power grids under continuously increasing EV penetration rates. But further research is needed to address robustness under extreme weather conditions. Conclusion: This study provides an effective load forecasting methodology for power systems to address the challenges of large-scale electric vehicle integration. Future research will explore more robust feature engineering methods and deep learning architectures, such as combining other more advanced time series prediction models and improving optimization algorithms to enhance model adaptability and generalization capability for complex data patterns.

摘要 : Expired fruit concentrates from agro-industrial processes, such as mango and pineapple, are typically discarded as waste, contributing to environmental pollution and economic losses. Valorizing these carbohydrate-rich wastes through biogas production presents a sustainable strategy for renewable energy generation and circular economy development. This study investigates the anaerobic digestion of expired mango and pineapple concentrates in a mesophilic continuous stirred tank reactor (CSTR) (Biogas Test Plant BTP2). The substrates were homogenized and co-digested with cow dung slurry as inoculum. Sample characterization (moisture, total solids (TS), volatile solids (VS), ash) followed APHA (2012) and ASTM D2974 protocols. Biogas production and composition (CH4, CO2) were monitored daily. VFA and TIC were determined through potentiometric titration according to Nordman's method. Microbial community profiling was conducted using rRNA gene sequencing . Principal component analysis (PCA) and Pearson correlation analysis were applied to identify the main operational drivers influencing methane yield. Results showed that cow dung (control) produced the highest cumulative methane yield (41.5 L' 68.7 % CH4), followed by mango (30.6 % CH4) and pineapple (28.5 % CH4). PCA revealed that pH stability, VFA accumulation, and stirring speed were significant drivers of methane production efficiency. Microbial analysis identified Firmicutes and Syntrophomonas as dominant contributors to methanogenesis across all reactors. This study innovatively combined PCA, correlation analysis, and microbial community profiling to evaluate biogas production from expired fruit concentrates in a CSTR system. The findings highlight the potential of these agro-industrial wastes for renewable energy generation and contribute to advancing sustainable waste management practices

摘要 : The catalyst materials 0N-Cu-MOF, 1N-Cu-MOF, and 2N-Cu-MOF were successfully synthesized usinga solvothermal method, and using different concentrations of nitrogen-modified Cu organic frameworks (xN-Cu-MOF). Characterizations using X-ray diffraction (XRD), scanning electron microscopy (SEM), fourier transform infrared (FT-IR) spectroscopy, X-ray photoelectron spectroscopy (XPS), and Brunauer-Emmett-Teller (BET) surface area analysis showed that 1N-Cu-MOF had the largest SSA and pore size among the three materials synthesized. 1N-Cu-MOF exhibited the largest pore size and specific surface area among the three materials, which had a decisive effect on CO2 reduction. In addition, stability and CO2 reduction reaction (CO2RR) activity were evaluated by linear sweep voltmeter, cyclic voltmeter, electrochemical impedance spectroscopy, and time flow tests. Faradaic efficiency (FE) was determined by product analysis. Among the three catalyst materials, 1N-Cu-MOF showed the best catalytic performance at 50 mA・cm-2 (maximum current density). The charge transfer resistance was 8.23 Ω, the average current density was 19.9 mA・cm-2, and the FE of methane (CH4) production showed a high efficiency of 70.45 % when tested for 12 h at an overpotential of -0.35 V (to-RHE).

摘要 : Sodium-ion batteries (SIBs) offer a viable alternative to partially or fully replace lithium ion batteries (LIBs) due to their lower cost and increased safety. This paper outlines the compositional optimizations, crystallographic evaluations, and electrochemical behavior of a novel mixed NASICON polyanionic compound, NaFe2PO4(SO4)2 (NFPS). X-ray photoelectron spectrometry (XPS) results showed that cobalt doping produces a higher concentration of oxygen defects compared to undoped samples. Scanning electron microscopy (SEM) analysis results revealed that the modified sample has more uniform pores and pore distribution. Brunauer-Emmett-Teller (BET) measurements showed that doping of Co2+ reduces the specific surface area of NFPS-Co0.08 compared to NFPS. This shortens the sodium ion diffusion pathway and promotes ion dynamics. The addition of Co2+ to the sample significantly improved its performance during galvanostatic charge-discharge tests. The electrochemical activity also is significantly enhanced by Co2+ doping. Na0.84Co0.08Fe2PO4(SO4)2 exhibits superior rate and cycling performance compared to pristine NFPS. After 80 cycles at 25 mA g-1, NFPS-Co0.08 retained discharge specific capacity of 60.8 mA h g-1, which is 1.24 times greater than that of NFPS.

摘要 : Collision avoidance between vehicles is a great challenge, especially in the context of mixed driving of connected and autonomous vehicles (CAVs) and human-driven vehicles (HVs). Advances in automation and connectivity technologies provide opportunities for CAVs to drive cooperatively. This paper proposes a proactive collision avoidance model, aiming to avoid collisions by controlling the speed and lane-changing behavior of CAVs. In the model, the subject vehicle first collects information about surrounding lanes and judges the traffic conditions; it then chooses to decelerate or change lanes to avoid collisions. The subject vehicle also searches for the optimal vehicle in the surrounding lanes for cooperation. The effectiveness of the proposed collision avoidance model is verified through the Python-SUMO platform. The experimental results show that the performance of the collision avoidance model is better than that of the cooperative adaptive cruise control (CACC) model in terms of average speed, lost time and the number of vehicle conflicts, proving the advantages of the proposed model in safety and efficiency.

摘要 : With the widespread application of multimedia data in cloud computing environments, ensuring the security and privacy of video content has become a pressing challenge. Traditional encryption methods have limitations; for instance, naive encryption algorithms and selective encryption can protect video content, but they struggle to balance codec compatibility and security. Full encryption algorithms have higher security compared to selective encryption, though they compromise time efficiency. Despite encryption techniques being employed to safeguard video content, there remain concerns about the security of encrypted videos, especially for highly confidential data. Furthermore, contemporary encryption methods often involve re-encoding and decoding processes, which adversely impact video processing speed. To address this issue, we propose a novel video security encryption scheme that optimally balances encryption speed and security. We introduce a scheme for video fragmentation, design corresponding sorting and merging algorithms, and construct a video transmission framework compatible with this scheme. Secondly, to validate the security of the scheme, we reduce its security to an NP problem and calculate the decryption time complexity, thereby demonstrating its security. Finally, experimental comparisons confirm the efficiency advantage of our scheme in processing large video files. Compared to traditional video encryption techniques, our scheme demonstrates faster overall processing speed and enhanced security when handling large video files. In summary, our scheme achieves a higher level of security while greatly preserving the efficiency of encryption and decryption, and the security foundation relies on cryptography.

摘要 : Microplastics, invasive species, and fungal pathogens increasingly threaten the sustainability of artificial grasslands. Their combined impact on alfalfa-based grassland remains poorly understood. This study assessed the individual and interactive impacts of microplastics, Solidago canadensis, and Rhizoctonia solani on growth, antioxidative mechanism, soil enzyme dynamics, and rhizospheric microbial communities. This experiment comprised of 4 alfalfa seedlings/pot, with/without amendments of 1% (w/w) microplastic/pot, application of R. solani and single seedling/pot of S. canadensis, both individually or in combination. Solidago canadensis significantly reduced plant height, leaf area and biomass by 57%, 48%, and 72% respectively. Root biomass was more vulnerable than shoot biomass, decreasing by 81%, 76%, and 78% due to microplastics, S. canadensis invasion, and R. solani, respectively. While catalase activity was higher in leaves, and that of peroxidase was greater in roots. However, activities of β-N-acetylglucosaminidase and β-Glucosidase increased by 80% and 70% respectively, under S. canadensis invasion, indicating intensified microbial processes. Sequencing of 16S rRNA yielded 1, 280, 487 reads, clustered into 17, 500 amplicon sequence variants with 122 common unique features, and maximum 3, 670 unique features were observed under microplastics. The relative abundance of top bacterial genera ( Limnobacter, Sphingomonas, Lysobacter, and Saccharimonadales ) was observed, and Chao1 and ACE indices were 683.37 and 682.33 for microplastics, respectively. PCoA explained 44.61% of total variation across PC1 and PC2, clearly separating treatments. Network correlation showed significant microbial associations shaped by stressors. These results demonstrate that microplastics, invasion, and pathogenic fungi significantly reduce alfalfa performance, alter rhizospheric enzymatic profiles, and reshape microbial community structures. Their interactions may threaten functionality and resilience of grassland ecosystems, underscoring the need for integrated strategies to mitigate multiple soil-based stressors.

摘要 : Distinct life history strategies of rhizosphere bacterial communities were established using the dilution-to-extinction approach. The life history strategies of rhizosphere bacterial communities influenced wheat growth during drought stress. Bacterial life history strategies exhibited a significant impact on community assembly in rhizosphere. A shift in rhizosphere bacterial life history strategies towards copiotrophy resulted in stable networks. The metabolic complexity of microorganisms can be simplified by classifying them into r-strategists and K-strategists. However, their associations with plant growth during drought remain largely unclear. Herein, we used the ribosomal RNA gene operon (rrn) copy number to characterize bacterial life-history strategies, with increased rrn copy numbers suggesting a shift from K- to r-strategies. We generated a series of bacterial communities with increased rrn copy numbers in rhizosphere. Drought decreased rhizosphere bacterial rrn copy numbers, rather than in root, indicating a prevalence of K-strategies during drought stress in rhizosphere. The rrn copy numbers of rhizosphere communities were negatively related to wheat growth during drought, while no significant associations were observed in control treatment. Rhizosphere bacterial communities with higher rrn copy numbers exhibited less community dissimilarity and tended to be more stable. Moreover, the abundance of most predicted functions decreased with rrn copy numbers in drought-stressed rhizosphere. Co-occurrence network analysis indicated that increased rrn copy numbers in rhizosphere community improved the proportion of negative to positive cohesion, implying more stable networks. Our findings bring up innovative knowledge about the relationships between microbial life-history strategies, communities and plant growth, and highlights the importance of plant-microorganism interactions for plant growth during stress.

摘要 : A random matrix is called a Gaussian matrix if its vectorization follows a Gaussian distribution. This paper investigates the covariances of random matrices and symmetric random matrices. We introduce the concepts of the Gaussian matrix, line normal matrix, and covariance tensors, and study the positive definiteness of the 4-order tensors. We show that the covariance tensors are positive definite, and characterize the covariance tensors of the standard normal distributed random matrices. We also use tensors to simplify the expression of the k-moments of random vectors and random matrices, and shows that all the k-moments of a centralized random vector with Gaussian distribution is zero if k is an odd number.

摘要 : Ancient architecture in Tibet has attracted considerable research attention because of its historical, religious, and artistic significance. Because the design and construction of ancient buildings often do not have the same detailed documentation as modern buildings, it is difficult to accurately simulate their structural properties. To study the structural performance of ancient buildings, the Tibetan ancient architecture structural performance analysis method based on finite element analysis is proposed. To better reflect the actual situation of engineering, the genetic algorithm is introduced to fine-tune the finite element calculation. The results of the experiments showed that the column compressive strain was maximum at 700 με . Moreover, the plastic strain occurred only when the compressive strain reached more than 1, 900, so the column was in the elastic strain stage during the process. When subjected to a vertical load, the upper part of the beam was compressed and the lower part was stretched. Furthermore, the lower part of the bow wood was stressed, and the upper part was stretched. The proposed method effectively improves the accuracy of the structural performance analysis of ancient buildings, and the results can be used for reference in ancient buildings.