Based on the development of Xinyang City, Henan Province, from 2010 to 2020, this study adopts a factor integration perspective. It constructs a multi-dimensional coupled system—comprising the Patch-generating Land Use Simulation (PLUS) model, a population flow model, a Principal Component Analysis-Back Propagation (PCA-BP) model for land scale prediction, and a GDP prediction model—to simulate the flow and integration patterns of key factors such as talent, capital, and land in 2035. Three scenarios were established: a baseline scenario, an economic priority development scenario, and an agricultural protection development scenario. The results indicate that: under the economic priority scenario, the proportion of urban land is the highest, the population is concentrated in urban areas, and GDP growth is the fastest. Under the agricultural protection scenario, cultivated land is effectively protected, the scope of population outflow is the most extensive, and economic growth is more balanced. Under the baseline scenario, urban-rural development is relatively moderate. The rational allocation of urban-rural factors is crucial to rural revitalization, and policy interventions significantly affect the direction and efficiency of factor flows. By employing machine learning methods to explore path selection and scenario simulation for rural revitalization, this research not only broadens the theoretical paradigms and policy support mechanisms for the field but also provides methodologies and findings that can contribute to the formulation of urban-rural development policies in other regions.

In recent decades, due to the intertwined effects of multiple factors such as global warming and intense human activities, the surface processes of global river basins have changed significantly. River geomorphologic changes are an important indicator of the long-term evolution of rivers. However, most current studies focus on a single aspect in river geomorphologic changes; systematic analysis of river dynamics from a macro perspective has not received enough attention. This study designed a new river geomorphologic change index, superimposing and analyzing the river changes in the Pearl River Basin from different periods, thereby revealing the spatiotemporal dynamics of the rivers in the Pearl River Basin. The study shows that: ① The stability of river geomorphology in the Pearl River Basin is relatively high, and the stability is gradually increasing with the intervention of human activities; ② Between 1995 and 2015, the Pearl River Basin experienced the greatest increase in average river area and the smallest decrease, indicating the most dynamic changes in river morphology during this period; ③ the stability of large rivers in the Pearl River Basin is higher than those of medium-sized rivers; ④ The stability in the middle and lower reaches of the Pearl River Basin is relatively high, but is low in the upper reaches. The Hejiang tributary basin reported the highest stability, while the Beipanjiang tributary basin reported the lowest stability. This study emphasizes the key role of retaining wide river corridors and restoring natural riverbed geomorphology for maintaining natural landforms, and puts forward suggestions for optimizing basin management and disaster prevention and mitigation.

Urban traffic congestion is one of the major challenges in modern cities. In urban traffic, drivers tend to choose the lane that seems the least congested at present because of limited vision, but this ″local optimal″ choice may lead to lower overall traffic efficiency. In order to solve this problem, this paper proposes an intelligent diversion system for urban highway and expressway. The system captures real-time video data through cameras in the traffic network, and uses Yolo and DeepSORT algorithm to accurately detect and track vehicles. These data are further mapped into a graph structure, and the graph neural network model is used to predict the lane-level traffic flow. Based on the forecast results, the system generates and updates the diversion information in real time, and recommends the best driving route to the driver through the road information board. The system has been deployed and performed reliably at the Chungang interchange section of South China Expressway in Guangzhou. Experimental data show that the system can improve the evenness of lane-level traffic flow, increase traffic efficiency, and verify the effectiveness of the intelligent diversion system.

The advancement of computer technology faces dual challenges posed by the physical limits of silicon-based chips and the energy efficiency bottleneck of von Neumann architectures, driving the shift toward emerging paradigms such as biomolecular computing. Deoxyribozyme (DNAzyme), as essential tools for biocomputation, have been largely confined to single-layer logic operations. This study proposes a modular molecular circuit design strategy by introducing allosteric regulation chains into the dynamic catalytic mechanism of DNAzyme, enabling the construction of programmable logic gates, including weighted-sum gates, threshold gates, and subtraction gates. Through cascading, molecular neurons and a molecular classifier were successfully implemented. This research, verified by NUPACK simulations and biochemical experiments, indicates that the designed logic gates can stably achieve specific computations. Furthermore, through Visual DSD simulations and experiments, the functions of molecular neurons and classifiers have been confirmed. This research provides a novel technical pathway for developing sophisticated biocomputing systems.

Hyperlipidemia, as a significant risk factor for atherosclerotic cardiovascular disease, has contributed to the increasing global burden of cardiovascular diseases, making it a major public health issue worldwide. Traditional lipid-lowering medications such as statins and fibrates have limitations, including side effects and poor tolerance, which makes the development of novel, safe, and effective drugs imperative. Peptide-based drugs, with their high bioavailability, potent activity, low toxicity, and ability to regulate lipid metabolism through multiple targets, have emerged as a new direction in lipid-lowering drug research. Lipid-lowering peptides are derived from a wide range of sources, including plants, animals, and microorganisms, and can be classified into natural and synthetic peptides. These peptides exert lipid-lowering effects through various mechanisms, such as inhibiting cholesterol synthesis, promoting lipid metabolism, and suppressing lipid absorption, with some peptides exhibiting multiple mechanisms of action. In recent years, various peptide drugs have entered the market, with PCSK9 inhibitors, such as evolocumab, demonstrating significant efficacy in clinical applications. Additionally, many novel therapeutic peptides are currently in preclinical and clinical development stages, showing promising prospects for future advancements.

In the multi-attribute group decision-making problem in the context of probabilistic interval-valued intuitionistic hesitant fuzzy number (PIVIHFN), the concept of weighted probabilistic interval-valued intuitionistic hesitant fuzzy set (WPIVIHFS) is proposed, in order to reflect the decision-making role of the expert in the whole process. Based on the theory knowledge of WPIVIHFS and TOPSIS idea, a decision algorithm of optimal scheme is set up. WPIVIHFS is described by six-dimensional point coordinates, such as median and clarity of interval-valued membership, median and clarity of interval-valued non-membership, probability and expert weight. Based on the description of WPIVIHFS in six-dimensional point coordinates, the correlation algorithm of weighted probabilistic interval-valued intuitionistic hesitant fuzzy element (WPIVIHFE) is established. The entropy method is used to calculate the objective weight of the attribute. Finally, the decision algorithm is applied to the problem of choosing the construction site of not-in-my-backyard facility. The research shows that WPIVIHFS contains the weight of review experts and contains higher-dimensional decision data information, so the decision algorithm based on WPIVIHFS has better applicability.

This paper improves the weak Galerkin generalized multiscale (WG-GMS) finite element method by integrating oversampling and online adaptive techniques to solve second-order elliptic equations with high-contrast coefficients. The oversampling technique reduces errors caused by boundary conditions by expanding the sampling region for the multiscale basis functions. Additionally, the online adaptive method constructs new online multiscale basis functions based on the maximum residual, thereby providing an improved function approximation space and improving the accuracy of the numerical solution. Numerical results demonstrate that the proposed oversampling and online adaptive mixed WG-GMS finite element methods are effective in solving second-order elliptic equations with high-contrast coefficients, achieving higher accuracy in the numerical solutions.

In order to achieve the testing and acceptance of the two state attitude integrated instrument, a testing system with graphic display and atmospheric calculation functions has been developed. Based on the requirements analysis, the overall architecture was determined, a display processing unit was designed, the current consumption acquisition schemes were compared, software was developed, and measurement interfaces were kept. Then, common faults and corresponding measures were summarized, and key difficulties were solved by establishing a fault tree. A standardized process was proposed for troubleshooting. Finally, the joint debugging effect between the comprehensive testing system and the tested instrument and its components was verified through experiments, and all functional items were tested to be qualified. The indicated airspeed error is close to 0 km/h, the pressure altitude error is less than 10m, and the lifting speed error is less than 0.4 m/s. The research results show that the development of the testing system is correct, and its performance and quality meet the expected standards, demonstrating strong practicality.

Counties are an important component of China's urban system and a key support for the integrated development of urban and rural areas. Changes in the number and structure of rural populations are of great significance for promoting the construction of new urbanization and rural revitalization. This paper takes the counties in Guangdong Province as the research object, uses statistical data such as the “Guangdong Rural Statistical Yearbook”, and employs methods such as multi-scale geographically weighted regression (MGWR) and geographical delection to study the spatio-temporal evolution of rural population distribution in Guangdong Province's counties from 2010 to 2020 and its influencing factors. The results show that: ① In terms of time, the changes in the rural population of the entire province present a dynamic trend of first increasing and then decreasing, which can be divided into a period of sustained growth (2010-2014) and a period of significant decrease (2014-2020); ② Spatially, the distribution of rural population density in Guangdong Province's counties generally shows a spatial pattern of “higher in the Pearl River Delta and eastern and western Guangdong, while lower in northern Guangdong”, and the decrease and growth of rural population present a spatial pattern of “mainly growth in the Pearl River Delta and decrease in eastern, western and northern Guangdong”; ③ The factors influencing the distribution of rural population are diverse. Per capita GDP and industrial structure upgrading have a negative effect on the distribution of rural population, while the level of medical services and the total value of agricultural production have a positive impact. The effects of labor productivity and grain output change over time; ④ The influencing factors of rural population distribution in counties show an enhancing effect through two-way interaction, and the level of medical services and the total value of agricultural production are the dominant interacting factors and show differentiation over time. This study not only enriches the theoretical and empirical research on the dynamic changes of the rural population in China, but also provides a scientific basis for promoting new urbanization and rural revitalization in a targeted manner.