Abstract:With the introduction of policies related to low-altitude economy,drone delivery, as an important driving force for enhancing new types of productivity,has gained widespread attention. Due to the limitations of drones in terms of batteries capacity and payload constraints, scholars have conducted extensive research on drone delivery models and systems. This paper aims to review the current literature in drone delivery and explore possible future research directions. This paper is conducted from three perspectives: the types of drone delivery problems,the modes of drone delivery systems,and the optimization scheduling methods for drone delivery systems. First, drone delivery problems are categorized into three types based on the services provided: pick-up only, delivery only, pick-up and delivery simultaneous, and point-to-point delivery. Second,drone delivery modes are divided into two categories based on whether drones are the sole tools for pickup and delivery:drone delivery systems and vehicle-drone collaborative delivery systems. Third, the paper introduces common optimization objectives in drone delivery problems, and reviews both single-objective and multi-objective optimization methods for solving drone delivery problems.

Abstract:The issue of excessive vibration in high-rise buildings located in the test track area of subway depots is studying. Through on-site measurement and finite element simulation of the environmental vibration before and after the vibration reduction transformation of a subway vehicle testing line, the vibration source and the vibration transmission law of the upper cover before and after the vibration reduction transformation were analyzed. The results indicated that the influence of vehicle speed on the dominant frequency of the source intensity affects the level of vibration transmitted to various floor slabs. The vibration measured in bedrooms was relatively large under the condition of 60 km/h, while the vibration in living rooms was more significant under the condition of 40 km/h. The vibration decreased first, then increased and then decreased with the increased of the distance between the measuring point and the vibration source. While the steel spring floating slab reduced vibration, noise may be slightly amplified, with all vibration measurement poited at the vibration source reduced by more than 10.00 dB after the modification, but the noise at the 7.5 m measurement point increased by 0.96 dB. By compared and observed the changes in vibration levels at different frequencies with floor height before and after the vibration reduction modification, it was found that the maximum frequency band of vibration acceleration level shifts forward, from 25~40 Hz before the modification to 20~25 Hz after, which was related to the change in excitation frequency at the vibration source. Compared to conventional ballast beds, the steel spring floating slab after vibration reduction modification can significantly reduce vibrations, enabling the buildings constructed above the test track area of this depot to meet the corresponding regulatory requirements.

Abstract:Investigation of the dynamic response of reinforced concrete columns under impact, relying on the two-degreeof-freedom mass-spring-damper model and combining with the OpenSees computing platform, on the basis of clarifying the equivalent resistance-displacement curve of reinforced concrete columns, a simplified impact analysis method for reinforced concrete columns integrating material nonlinearity and impact strain rate effects is proposed. By comparing with the drop hammer impact test of reinforced concrete columns, the validity of the proposed two-degree-of-freedom simplified analysis method is verified. On this basis, the impacts of sensitive parameters like impact velocity, impact mass, axial compression ratio and longitudinal reinforcement ratio on the drop hammer impact response of reinforced concrete columns are explored. The results showed that the peak impact force calculated by the proposed simplified analysis method has a deviation of approximately 4.30% from the test value, and the deviation of the peak displacement at mid-span is approximately 2.16%. With the increase of the impact velocity and impact mass, the peak impact force and the peak displacement at the mid-span of the column gradually increased, but the increase of the impact velocity and impact mass would lead to a de-layed characteristic of the peak displacement at the mid-span of the column. With the increase of the axial compression ratio and the longitudinal reinforcement ratio, the peak impact force of the column gradually increased, while the peak displacement at the mid-span showed a gradually decreasing trend. Moreover, the change of the longitudinal reinforcement ratio had a significantly greater influence on the secondary impact in the impact force time history curve than the change of the axial compression ratio.

Abstract:To reveal the reinforcement mechanism of pipe shed support on the tunnel face of soft rock tunnel and evaluate the stability of the tunnel face under pipe shed support, the numerical simulation method is used to compare and analyze the failure mode of tunnel face with or without pipe shed support. Then, based on the numerical simulation results, a new parabolic arch-logarithmic spiral tunnel face failure model considering pipe shed support and soil arch effect was proposed, and the calculation method of tunnel face safety factor under pipe shed support was derived by upper bound limit analysis method and strength reduction method. Finally, the influence of surrounding rock parameters and pipe shed support parameters on the stability of tunnel face was discussed. The research results indicate that when the instability failure of the tunnel face without pipe shed support occurs, the surrounding rock above the tunnel face forms a large collapse arch and a small collapse arch. Because of the pipe shed support, the instability failure of the tunnel face dose not develop to the vault. Meanwhile the pipe shed support prevents the formation of local collapse arch above the unsupported section, and makes the displacement direction of the surrounding rock above the unsupported section change from vertical downward to extru-sion to the tunnel face. The safety factor of tunnel face stability increases with the increase of internal friction angle, cohesion of surrounding rock and the size of pipe shed, and increases with the increase of single excavation length and pipe shed spacing. Too large size of the pipe shed is not economical for enhancing the stability of the tunnel face.

Abstract:In order to explore whether time-dependent fare strategy can be introduced into the subway operation of type I big cities to alleviate the severe passenger flow congestion brought by holidays, the survey data of Nanchang subway respondents is used to analyze the travel time choice behavior of subway passengers in type I big cities. SP survey method was used to collect data on respondents’travel time choice preferences under the influence of three scheme variables: travel time change, fare and congestion level. Conditional Logit model and mixed Logit model were used to estimate the variable coefficients of utility function. The results show that the change length of travel time has a significant and negative effect on travel time choice behavior, and the fare level has a significant and positive effect on travel time choice behavior. Compared with the change length of travel time and the congestion level, the fare has a greater impact on the travel time choice behavior, and the respondents are more willing to choose the pre-peak period. The research can provide decision-making reference for relieving passenger flow congestion of subway in type I big cities during holidays.

Abstract:The ring roads in mountainous city business districts are prone to congestion due to numerous intersections and fluctuating traffic flows. To address this, a cooperative signal control method for managing ring roads entrances and exits is proposed. Road and traffic characteristics were extracted using high-altitude video, checkpoint data, and Internet-based datasets. The control objectives include maximizing vehicle throughput, minimizing entrance queue lengths, and reducing downstream congestion at exit intersections.The optimization model is constrained by signal control parameters, queue lengths, and overall ring roads density. The standard NSGA-Ⅱ algorithm was improved in operator design and population size configuration. A MATLAB-VISSIM simulation platform was developed for joint simulation. A case study of the Guanyinqiao business district ring roads in Chongqing was conducted. Results show that, compared to the original scheme, the optimized strategy increases ring roads capacity by 8.9% during peak hours, reduces entrance lane queue length by 31.7%, and decreases exit lane queue length by 15.0%. These findings validate the effectiveness of the proposed multi-objective optimization and cooperative signal control method. By enhancing coordination among multiple intersections, this approach provides valuable guidance for alleviating regional traffic congestion during peak hours.

Abstract:To achieve accurate prediction of the flight punctuality rate, a flight regularity prediction index system was constructed based on data statistics of flight delay reasons, includeing departure airport, destination airport, flow control information, and route characteristics. It proposes a SMOTE algorithm-based XGBoost classification prediction model (SM-XGBoost model) and a SMOTE algorithm-based LightGBM classification prediction model (SM-LightGBM model). Based on the actual data of major airports in East China, the validity and progressiveness of the proposed model are verified. The results showed that the SM-XGBoost model and SM-LightGBM model were significantly better than the decision tree and random forest models in terms of prediction accuracy and error. In terms of stability of training set and test set, SM-LightGBM model is superior to the SM-XGBoost model, with a maximum prediction accuracy of 88.2% for test set. This method provides a new analytical approach for predicting events in similar complex systems.

Abstract:This study uses the Guangzhou terminal area and the Zhuhai terminal area as examples, applying complex network methods to construct route network models for each. The network topology is analyzed in detail using indicators such as node degree, degree distribution, betweenness centrality, clustering coefficient, network diameter, and average path length. Based on this analysis, node importance is identified using degree centrality, betweenness centrality, and the KBKNR algorithm. Additionally, scenarios of random attacks and deliberate destruction are designed, and network efficiency and connectivity are used as metrics to evaluate the resilience of the route networks.The experimental results show that under random attacks, the Zhuhai terminal air route network demonstrates greater resilience compared to the Guangzhou terminal air route network. Conversely, under intentional attacks, the Guangzhou terminal air route network exhibits stronger resilience. Moreover, degree centrality attacks are more likely to cause network collapse compared to betweenness centrality attacks and KBKNR attacks, indicating that nodes with high degree centrality should be prioritized for protection as critical nodes.

Abstract:Fully exploiting the spatial correlation of passenger flow between related stations in the subway network has a positive effect on the improvement of subway passenger flow prediction accuracy. Capturing and quantifying spatial patterns in passenger flow data is difficult due to the difficulty of learning and transferring spatial correlations between metro stations. An improved graph-convolution gated recurrent neural network (GCGRU) metro passenger flow prediction model was proposed to enhance the model’s ability to handle different data types by integrating multivariate spatiotemporal data. The spider wasp optimisation (SWO) algorithm based on Tent chaotic mapping and Levy flight disturbance strategy was used to dynamically adjust the model structural parameters in order to optimize the hidden layer structure of the gated recurrent neural network. The experimental results show that the prediction accuracy of the model is significantly higher on weekdays than on weekends, and the root mean square error, mean absolute error, and mean absolute percentage error are reduced by 13 percentage points, 12 percentage points, and 0.08 percentage points, respectively, during weekdays compared to weekends. Dynamic optimization of the hidden structure of gated recurrent networks can lead to better convergence of the prediction model and higher prediction accuracy

Abstract:The influence of high-speed railway on urban green technology innovation represents a pivotal area of study within the broader research on the impact of high-speed railway on regional innovation. To investigate the effect of high-speed railway on the“dual enhancement of quantity and quality”in urban green technology innovation, a multi-period Difference-inDifferences (DID) model was established. This model utilises panel data from 41 cities within the Yangtze River Delta urban agglomeration spanning the years 2004 to 2019 to explore how the high-speed railway opening influences urban green technology innovation. The findings reveal that the introduction of high-speed railway significantly promotes the“dual enhancement of quantity and quality”in green technology innovation across the Yangtze River Delta cities. Mechanism analysis further indicates that high-speed railway impacts urban green technology innovation by bolstering economic agglomeration and venture capital investment. Additionally, the effects of high-speed railway opening exhibit heterogeneity in terms of city size and regional characteristics, with notable impacts observed in both large cities and small to medium-sized cities. Specifically, the influence of high-speed railway is more pronounced in Jiangsu Province compared to Zhejiang and Anhui Provinces. This study systematically elucidates the relationship between the high-speed railway opening and the“dual enhancement of quantity and quality”in urban green technology innovation within the Yangtze River Delta region. The conclusions remain robust following rigorous stability tests.

Abstract:Deep learning technology offers advantages in vibration signal recognition with high accuracy and precision. However, acquiring a large number of labeled data for polygonal wheel detection is challenging, which fails to meet the training requirements of conventional neural network models. Existing methods to address the issue of small sample sizes often convert time-domain data into frequency-domain data, but this can re-sult in the loss of certain data features during the time-frequency conversion. To address this issue, a polygonal wheel detection method based on the 1DResAE deep neural network model is proposed. This model completes the detection of polygonal train wheels by unsupervised learning, feature extraction, and supervised learning of time-domain signals without the need for time-frequency conversion of vibration signals. By integrating one-dimensional convolution, residual networks, and autoencoders, a one-dimensional deep neural network is formed, capable of extracting and learning complex one-dimensional vibration signal features. Based on the features extracted and learned by the encoder in the autoencoder, the classifier performs supervised learning with a small amount of labeled data to achieve pattern recognition of polygonal train wheels. Experimental verification using data collected from a small-scale wheel-rail rolling test bench demonstrated that the detection accuracy of this method is 98.971%, with low error and outstanding classification performance. For the task of polygonal wheel detection, the 1DResAE model effectively detects the polygonal order of wheels and has practical applicability.

Abstract:The BP neural network improved by the dung beetle optimization(DBO) is used to calculate the stress concentration factor (SCF) of T-tubular joints, and the SCF can be solved quickly and accurately. First, finite element parameterized modeling of T-tubular joints under basic axial loading was conducted, and comparative analysis with experimental data verified the model’s reliability. Next, a SCF dataset was established for crown and saddle points, analyzing the influence of dimensionless geometric parameters on SCF. Finally, the BP neural network improved by DBO is used to perform regression prediction on the SCF data sets of joints with different geometric parameters. The results show that the prediction performance of the improved BP neural network model is better than that of the unimproved BP neural network. Compared with the SCF parameter equation, the BP neural network prediction using DBO is more efficient and accurate.

Abstract:To address the limitations in detection accuracy and inference speed in current road crack detection models, this paper proposes a novel YOLOv8-Crack network model. Based on YOLOv8n, this model incorporates multiple key structural optimizations, including the introduction of the NWD loss function to reduce dependency on aspect ratios of bounding boxes, thus improving detection capability for irregularly shaped cracks. The Slimneck lightweight structure is used to significantly reduce the number of parameters and computational complexity of the model, and accelerate the inference speed. The model also integrates a CA module to enhance the capture of critical feature information. Experimental results on the open-source dataset RDD2022 demonstrate that the YOLOv8-Crack model achieves improvements over the original YOLOv8n, with precision, recall, and mean average precision increased by 1.8%, 3.7%, and 2.6%; respectively, while parameters and computation are reduced by 6.7% and 11.0%.