In order to study the mechanical properties of Changxin epoxy binder used in Yangjiang Port Bridge, its performance under tensile strength, tensile strength, shear strength and other conditions were evaluated through a series of experimental and theoretical analyses. The research includes the preparation process of binder, mechanical property testing, failure analysis and performance comparison with Japan TAF epoxy binder. The results show that Changxin epoxy binder exhibits excellent mechanical properties and stability in steel bridge deck pavement, which provides a strong guarantee for the long-term use of Yangjiang Port Bridge.

In order to explore the mechanical characteristics of shield tunnels passing through different soft-hard uneven strata, a three-dimensional refined shield tunnel numerical model considering the contact effect of ring longitudinal bolt joints and segment-soil was established, and its feasibility were verified by comparing the measured data. Then, aiming at the complex ground conditions of upper soft-lower hard, upper hard-lower soft, and longitudinal soft-hard uneven strata, the influence of different strata composite ratios on the internal force, convergence deformation and vertical settlement of shield segment structure was deeply explored. The results show that with the increase of strata composite ratio, the bending moment of tunnel bottom in upper soft-lower hard strata decreases greatly, the vertical settlement of segment increases slowly and then rapidly, and the convergence deformation of segment increases linearly. Compared with the upper soft-lower hard strata, the vertical settlement response of shield segment in the upper hard-lower soft strata is more sensitive under the same composite ratio. When the tunnel lies in the longitudinal soft-hard uneven strata, the extreme value and extreme difference of the longitudinal bending moment of the segment and the shear force increase with the increase of the strata composite ratio. The convergence deformation of the segment is larger on the hard soil side than on the soft soil side.

Ground penetrating radar(GPR) is widely used for inspecting the quality of tunnel linings. However, the raw GPR data often cannot be directly interpreted and requires various pre-processing such as denoising, gain adjustment, and image smoothing to observe meaningful information. Considering that GPR data processing is currently predominantly manual, with a complex workflow and subjective parameter selection, an end-to-end data processing method is proposed based on generative adversarial network(GAN) that transforms raw GPR data into images with clear signals. The GAN consists of a series of generators and discriminators at different scales, capable of intelligently recognizing both global and local features of GPR data and automatically performing comprehensive processing operations on the raw data. This method has been successfully applied to the processing of actual GPR data for initial lining quality inspection, achieving results comparable to manual processing and a significantly higher data processing efficiency.

Under the technical support of geographic information system(GIS), the country park in Yannan Mountain Area of Urumqi was taken as the research object. Based on the field investigation, 10 key ecological sensitivity evaluation indexes were selected. By applying analytic hierarchy process (AHP) to determine the weight of each factor, the evaluation index system of ecological sensitivity of country parks was constructed. The results show that the ecological sensitivity of country parksis mainly highly sensitive, with an area of 173.30 hm², accounting for 31.64%. The remaining sensitive areas are 91.07 hm², 93.01 hm², 126.07 hm², and 64.30 hm², accounting for 16.63%, 16.98%, 23.02%, and 11.74%, respectively.

Addressing the difficulties in collecting key data during battery operation and the limited amount of electrochemical impedance spectroscopy (EIS) data, is able to optimize battery performance evaluation and health monitoring, as well as optimize battery usage and charging strategies. The research method involves using data augmentation techniques to increase the sample size while ensuring data quality. The denoising diffusion probability model (DDPM), as an emerging generative model, is applied to enhance battery data. For low dimensional battery data such as current, voltage, temperature, and capacity, the DDPM model is directly applied for data augmentation. For high-dimensional EIS data, the autoencoder (AE) model is first used for dimensionality reduction, followed by data augmentation in low dimensional space, and the enhanced data is restored to the original space. The research results confirm that the proposed data augmentation method can generate high-quality data on NASA(National Aeronautics and Space Administration) and EIS public datasets and effectively reduce computational complexity. The conclusion indicates that this study provides an effective data augmentation strategy for battery performance evaluation and health management, and has certain reference and application value.

Taking Qianguan Bridge in Dalian as the project background, the structural characteristics and stress conditions of the lower-supported steel pipe concrete tied arch bridge were analyzed. A health monitoring system scheme for the bridge was designed. The architecture design of the system and the types of sensors used for its main monitoring content were introduced. Finite element analysis of the bridge was conducted based on Midas Civil, and the measurement point layout of the bridge monitoring system was determined in accordance with relevant standards. Considering the strong subjectivity and poor overall nature of traditional manual inspections, which consume a lot of manpower, material resources, and financial resources on-site and even affect traffic operations, an intelligent bridge health monitoring platform was developed in the cloud based on BIM(building information modeling). This platform integrates real-time on-site data collection, stable and efficient transmission, rapid analysis of massive data, human-computer interaction, and visualization of models and data. It automatically triggers alerts and notifies relevant personnel through various means, achieving early detection, early warning, and early handling, providing strong support for the ultimate realization of smart bridge management and maintenance.

With the rapid growth of the number of new energy vehicles,as an important infrastructure, the location of battery swap stations is crucial to improve the user experience and promote the popularization of electric vehicles. In order to explore the location strategy of new energy vehicle battery swap station with the goal of maximizing user satisfaction, the key factors affecting user satisfaction, including service cost, travel time, waiting time and power consumption were analyzed through the comprehensive consideration of these four factors. The multi-objective decision-making problem is transformed into a single-objective optimization problem according to the CRITIC weight method, and the site selection optimization model was constructed, and the genetic algorithm was used to analyze the example to obtain the optimal location location.

The aviation material supply chain is an important part of ensuring the normal operation and punctual operation of civilian aircraft. A supply chain stability model based on pulses and uses MATLAB for numerical verification was constructed to explore the role of pulse intensity and occurrence in the stability of the supply chain. The research results show that pulses have a greater impact on supply chains that are not yet stable, and a limited impact on supply chains that are already stable. Appropriate pulses accelerate system adjustments and quickly reach a stable state.

Aiming at the airport clear water vehicle scheduling problem, considering the situation that clear water vehicles with multi-specification load capacity provide guarantee services for flights of different aircraft types, an optimal airport clear water vehicle scheduling model was constructed to minimize the number of vehicles in use and the waiting service time for flights. By introducing time window span and service waiting time as the key factors of the state transfer rule, and adopting stage pheromone evaporation and intra-path 2-opt and inter-path 2-opt^* optimization strategies, the traditional ant colony algorithm is improved in terms of applicability and used for model solving. Finally, the operational data of a hub airport in Southwest China was used as an example for validation. Compared with the traditional first-come-first-served scheduling, the number of vehicles in use and the flight waiting service time are reduced by 27.3% and 22.9%, respectively, for the scheduling scheme obtained by solving the improved ant colony algorithm. The results show that the constructed model and algorithm demonstrate high optimization efficiency and practicality in the airport clear water vehicle scheduling problem.

Agricultural high-quality development (AHQ) is an urgent need for the comprehensive promotion of rural revitalization today.Benchmark regression and spatial Durbin models were employed to investigate the effects of the coupling degrees of three sets of binary and ternary relationships-formed by ASTR allocation, AST innovation, and AHQ-on AHQ across each province from 2010 to 2020.The findings indicate that, overall, the coupling development among these systems plays a significant role in promoting AHQ. Specifically, the coupling degrees of both the binary and ternary systems markedly boost AHQ. However, when analyzed regionally, the degree of system coupling development exerts varying impacts on AHQ across the four major economic regions. Spatial econometric regression results further reveal that the coefficients for the impact of system coupling on AHQ, along with the spatial lag coefficients, are positive. This implies that the synchronized development of these systems not only positively affects local AHQ but also catalyzes AHQ improvements in neighboring regions. Therefore, this study suggests fully leveraging the unique strengths of each region in the integration of science, technology, and the economy, accelerating new quality agricultural productivity, capitalizing on the spatial spillover effects of system interactions on AHQ in adjacent areas, intensifying supply-side structural reforms in the agricultural science and technology sector, and enhancing the efficiency of ASTR allocation.