Efficient and accurate crack detection can provide a basis for assessing the structural safety of tunnels. Aiming at the shortcomings of traditional crack detection methods, which are complex and weak in generalization ability, an improved algorithm YOLOv5-CT(YOLOv5 CBAM Transformer) for tunnel lining crack detection was proposed.Considering the slender morphology of the cracks, the network introduced the Transformer module to improve the crack detection effect.The strong long-range dependency capture ability of the Transformer module enabled the proposed detection model to fully learn the contextual information of the crack region. In addition, the network integrated the convolutional attention mechanism CBAM(convolutional block attention module) in neck.The experiment shows that the YOLOv5-CT can achieve AP50 and AP of 85.2% and 51.3%, respectively, which is an improvement of 8.9% and 12.1% compared to the baseline model YOLOv5. It is better than other one-stage object detection networks in terms of accuracy, and the inference speed reaches 161.3 fps under 640×640 pixel conditions, which meets real-time detection of tunnel lining cracks.

In recent years, spatial-temporal graph convolutional network (STGCN) has been introduced into traffic flow prediction, which has good spatial-temporal traffic data modeling ability and has achieved advanced performance, but there are still two problems: ①Traffic flow data have strong temporal and spatial correlation; ②Static pre-defined graphs are difficult to capture the spatio-temporal dependence of dynamic changes in traffic flow over time. To solve the above problems, a new spatial-temporal decomposed framework (STDF) was proposed, which used residual connection, forgetting gate and update gate to organically connect time module and space module to decompose and predict input information in multiple dimensions. In addition, by instantiating STDF, a new traffic prediction model based on input traffic signal decomposition decomposed dynamic spatial-temporal graph convolutional network (DDSTGCN) was proposed. It captured the spatiotemporal dependencies of traffic and designed a dynamic graph learning module that takes into account the dynamic nature of spatial dependencies. Finally, two real traffic flow data were used to compare with the existing traffic flow prediction algorithms. The experimental results show that the proposed method has good performance in the accuracy of traffic flow prediction and can effectively complete the traffic flow prediction in the real scenario.

The evaluation of on-street parking efficiency is considered a necessary basis for the optimization and management of on-street parking as well as the development of urban intelligent parking systems. Therefore, an evaluation method of on-street parking efficiency based on the entropy weight Critic-Topsis model was proposed. Firstly, nine indexes, such as the number of parking spaces, the remaining width of the road, and the cumulative number of parking, were selected from three aspects: parking space attributes, operating characteristics, and traffic impact, to construct the evaluation index system of on-street parking efficiency. Secondly, an evaluation method of the entropy weight Critic-Topsis model was proposed based on the data characteristics of the evaluation index of on-street parking efficiency. The entropy weight method and Critic method were employed to calculate the weight of each index, and then the improved topsis model was utilized to quantitatively evaluate the parking lots. Finally, this model was utilized to analyze the data of 10 on-street parking lots in Beijing, resulting in the evaluation results and rankings for each parking lot. It is found that in the evaluation index system of on-street parking efficiency, the number of parking spaces, the utilization rate of parking spaces, and the remaining width of the road have significant influences on the evaluation results, with weights of 22.99%, 13.72%, and 13.71% respectively. In the case analysis of 10 on-street parking lots in Beijing, higher scores are observed for on-street parking lots such as Gongti East Road, Keyuan Road, and Huayuan North Road, which are 0.659 4, 0.611 3, and 0.608 1, respectively. Moreover, relevant suggestions and optimization measures are provided for the on-street parking lots of Chunhe Road and Lingnan Road, which have lower scores, aiming to enhance the long-term management system of urban on-street parking and improve the level of urban traffic management.

The selection of variables affecting fuel consumption in the existing studies usually has no clear criteria, and it is difficult to combine the research results with actual flight. The flight training data of a Cessna 172 was used to predict the fuel consumption during the airborne phase of general aviation trainer aircraft. Firstly, based on the authors’ flight experience as well as correlation analysis, the features that influence fuel flow rate were selected from the pilot’s operational perspective. Secondly, a regression tree model was used to predict fuel flow rate under different flight conditions, correlating the aircraft’s actual flight status with the predicted fuel flow rate, in order to facilitate subsequent research on specific fuel-saving strategies from the flight technique perspective. Finally, a random forest model optimized with hyperparameter tuning was used to predict the fuel flow rate. The experimental results show that the accuracy of the model used is better than that of the existing research results, with a mean absolute error of 0.286 gallon/h, a root mean squared error of 0.496 gallon/h, a residual sum of squares of 0.968 4, and a mean absolute percentage error of 4.00%.

In order to solve the problem of uneven allocation of airspace resources in traditional artificial sectors based on subjective experience, and to meet the needs of today’s air traffic operation, the problem of three-dimensional sectorization in terminal areas was studied by improving Agent method. Firstly, while adhering to traditional sectoring constraints, the objective was to enhance sector adaptability to traffic flows and achieve a reduction and balance in air traffic control workload. Subsequently, the traditional Agent method was improved by using genetic algorithm to determine the location of Agent initial solution, so that it could enhance computational efficiency, designing and optimizing Agent growth rules and spatial filling rules. Finally, using the Shanghai terminal area as a case study, the results indicated that the improved Agent method yields sector planning scheme with respective improvements of 25.84% and 18.54% in sector shape characteristics and adaptability to airborne traffic flows. Simultaneously, while reducing the overall terminal area air traffic control workload, the standard deviation of control workload among sectors was reduced by 53.33% and 36.58%, respectively, compared to the existing and traditional Agent methods.It can be seen that the Research on Improved Agent-Based Sectorization Method provides reference for the local characteristic airspace planning of our country.

Under the condition of airport autonomous operation, perception of the operational environment is a crucial factor constraining the realization of autonomous airport operations. In the process of airport surface traffic operation, understanding the utilization of surface movement resources is a key step in establishing a comprehensive operational environment. The surface movement process at airports is first focused on in this study, and an ontology model for airport surface movement processes is constructed. Based on the structural layout of the airport surface road network, the movement paths were divided, and a "node-edge" model based on the connection between network nodes was established. Meanwhile, building upon the ontology model, dynamic and static attributes of the surface road network were defined as the basic properties of network nodes. With network nodes as the research object, various conflict scenarios existing in aircraft surface movement processes were modeled based on the dynamic attributes of network nodes, thus achieving a dynamic representation of aircraft movement processes at network nodes. Using speed data generated by aircraft dynamics models as a basis, a visualization representation of dynamic graphs of surface movement resource utilization in the presence of aircraft conflict scenarios was designed.Experimental results demonstrate that the model effectively represents both conflict and conflict-free scenarios in surface operations. This enhances the overall perception of surface movement resource utilization among participants in airport surface traffic.

As a key device for rocket launch, the erecting device’s load-bearing performance is crucial for the success. In response to the instability phenomenon of the vertical plate of a certain rocket erecting device under the loading condition at the moment of erection, in order to analyze the reasons for the instability of the vertical plate, a finite element simulation model of the erecting device was established considering the actual load situation. The compressive load on both sides of the vertical plate under the working condition of erection was extracted, and based on the theory of small deflection thin plate elastic stability, the local instability of the vertical plate was explained: the compressive load of the vertical plate at the moment of erection exceeded its critical instability load, manifested as the characteristic of lateral bending deformation instability. Based on the analysis results, local reinforcement measures for the vertical plate were proposed. Finite element analysis and erecting loading tests were conducted on the vertical device after reinforcement. The results indicate that the stress consistency at the corresponding measurement points of the left and right vertical plates is good, and the stress deviation between the simulation calculation and the test result is not more than 10%. The lateral bending deformation decreases from 5.3 mm before reinforcement to 1.6 mm after reinforcement, proving the effectiveness of structural reinforcement. The relevant conclusions provide a reference for the local stability analysis of large and complex structures.

In view of the problems of large lag and nonlinearity in aeration control systems for wastewater treatment. The principles of aeration control systems were analyzed, meanwhile, mathematical model for such systems was established. Based on traditional PID(proportion,integration,differential) control algorithms, particle swarm optimization algorithms, and fuzzy control algorithms, an improved particle swarm optimized fuzzy PID algorithm was proposed to overcome the drawbacks of expert-dependency and lack of dynamic performance in fuzzy PID control. The system was simulated using MATLAB to compare the speed, accuracy, and stability of the three control methods in terms of step response, disturbance rejection, and robustness under model mismatch conditions. The results indicate that the improved particle swarm optimized fuzzy PID algorithm outperforms traditional PID and fuzzy PID control algorithms in terms of step response, disturbance rejection, and robustness. It achieves faster and more stable regulation of dissolved oxygen, thereby enhancing control system performance. The improvement is expected to reduce operational costs at wastewater treatment plants, as well as improve system reliability and economic efficiency.

To address the fuzziness and randomness in safety evaluations of liquefied natural gas (LNG) chemical plants, a novel risk assessment method was proposed based on the normal cloud model. The proposed method was based on the selection of evaluation indicators from the perspective of intrinsic safety, which was used to evaluate the severity and likelihood of accidents through expert judgment. The weights of the indicators were determined by combining the analytic hierarchy process (AHP) and the criteria importance though intercrieria correlation (CRITIC). The risk matrix thresholds were softened using a forward cloud generator, and the evaluation results were optimized through a backward cloud generator to obtain cloud characteristic numbers and generate corresponding two-dimensional cloud maps. By comparing the actual cloud map with the standard cloud map, the risk levels of the evaluation indicators were determined. The results show that the model effectively integrates the expert opinions with aims of retaining the randomness and fuzziness, and providing the visualizes the risk assessment results. The novel method also provides a new risk assessment tool for the safety management of LNG chemical plants.

Laser 3D scanning technology rapidly acquires point cloud data of target surfaces, including spatial point coordinates that describe the geometric features of the target and laser reflectance intensity that characterizes the material’s reflectivity. The application of automatic semantic segmentation techniques for 3D point clouds in geological exploration research lays the foundation for depicting regional geological features. To demonstrate the recent advancements of 3D laser scanning technology in large-scale semantic segmentation within geological scenarios, firstly, photogrammetry and LiDAR as two methods for acquiring 3D point clouds were compared, highlighting the advantages of LiDAR in terms of accuracy, versatility, and insensitivity to lighting conditions. By elucidating the principles of lithological semantic segmentation, a comprehensive review and summary of recent methods based on geometric or intensity features were provided. Common large-scale point cloud datasets and evaluation metrics were introduced, and the segmentation performance of different algorithms was compared. Finally, the limitations of existing methods were summarized, and future research directions for lithological semantic segmentation tasks were outlined.