In order to better analyze the stable sub-safety state and unstable sub-safety state existing in control sector operation，K-means algorithm was used to cluster three control sector operation stability evaluation indicators of excess-capacity ratio(ECR)，retention degree and flight attitude mixing ratio so that the optimal classification of the operation stability of control sector was determined. The index threshold corresponding to each level was obtained by clustering analysis of a single index. Combined with the index weight calculated by the entropy weight method，the operational stability level of the control sector in each time period was obtained according to the principle of maximum membership degree. Then，the comprehensive evaluation model of control sector operation stability was constructed. The actual flight data of Xiamen No.01 sector was selected to more comprehensively analyze operation situation of control sector from the perspectives of stability and trend. The results show that the best effect is obtained when the control sector operation stability level is divided into three categories. The stability varies with time due to the influences of air traffic flow and control conditions，especially in the two time periods of 7: 30-9: 15 and 20: 00-21: 00，when the change of stability is most obvious. The controllers need to pay great attention to improving the safety of airspace operations.

To ensure the SOTIF of eVTOL vehicles in UAM and reduce the verification difficulty of artificial intelligence algorithms，an obstacle avoidance model was proposed based on RTA method. Firstly，SAC (soft actor-critic) algorithm improved by the artificial potential field method was used as the complex function of the eVTOL intelligent obstacle avoidance system. Then，dynamic response planning (DRP) was used as a backup function of the intelligent avionics system to mitigate SOTIF hazards. Moreover，monitoring and decision-making modules were adopted to obtain environmental conditions and develop an RTA architecture. Finally，the simulated obstacle avoidance performance was compared between the two systems using complex function and RTA. The results showed that both methods can achieve obstacle avoidance，but the traditional obstacle avoidance system using complex functions may impose SOTIF risk. The RAT architecture design increased safe flight time from 78.4% to 98.15%，with the total route length only increasing by 0.95%，reducing risks in operational scenarios while ensuring efficiency.

To mitigate the risks of leakage，fires and explosions in petrochemical equipment，focusing on a typical catalytic cracking unit，a novel early warning method for detecting abnormal states using probability distribution functions was introduced. Spline fitting principles were used to uncover the trends in operating parameters such as pressure，temperature and flow rate over time，and to extract characteristic parameters such as deviation rate and deviation amount. By employing the Weibull distribution，the failure probability distribution function of the equipment was determined. The extracted characteristic parameters were integrated with the failure function to construct a probabilistic distribution mathematical model incorporating these features. Based on this model，a comprehensive early warning process was developed，facilitating real-time risk assessment and anomaly detection during the catalytic cracking process. The findings demonstrate that this method can effectively predict anomalies under conditions of oscillation，step changes，and gradual trends in operating parameters. Compared to traditional instrument systems，this early warning method advances the warning time by 87 to 621 seconds，addressing the limitation of limited response time following single-threshold alarms in the conventional systems. Furthermore，a comparison of various data processing methods reveals that the early warning model based on spline fitting exhibits superior performance.

To solve the problem that traditional safety design methods based on manual inspection were difficult to cope with the explosion of optional residence solutions caused by the large-scale integration of avionics systems，an avionics system partition model，task model and safety criticality level quantification model were constructed，and the comprehensive design optimization considering safety was modeled as an MDP problem. An optimization method of Soft Action-Critic (SAC) algorithm based on Actor-Critic framework was proposed. In order to obtain the correlation between the parameter selection and training results of SAC algorithm，the sensitivity of the algorithm parameters was studied. At the same time，to verify the superiority of the optimization method based on the SAC algorithm in optimizing the comprehensive design considering safety，optimization comparison experiments were carried out with the Deep Deterministic Policy Gradient (DDPG) algorithm and the traditional allocation algorithm as the objects. The results show that under the optimal parameter combination，the maximum reward after using convergence of SAC algorithm increases by nearly 8% compared with other parameter combinations，and the convergence time is shortened by nearly 16.6%. Compared with the DDPG algorithm and the traditional allocation algorithm，the optimization method based on SAC algorithm has improved approximately 62%，7464%，8370%，2123% and 775% in terms of the maximum reward，cumulative constraint violation rate，partition balance risk effect，partition resource utilization and solution time

In order to improve the science and effectiveness of traceability and localization of hazardous gas leaks，determining the location and intensity of dangerous gas leaks is the key to emergency response to accidents. The Gaussian plume model was modified by analyzing the mass conservation law and improving the diffusion amplitude of the gas plume with an approximate Gaussian distribution. Additionally，a heuristic algorithm based on the principle of immunization—IA coupled with PSO—was proposed，and the PSO-IA algorithm was applied to source strength inversion. It is concluded that the modified Gaussian plume model has been verified by three classical algorithms (PS，GA and PSO)，resulting in a prediction value error decreased by about 2%. PSO algorithm，which showed a better inversion effect，was selected for comparison with the PSO-IA algorithm. The PSO-IA algorithm has improved the effect of inverting source strength，with a localization error is 1.3 m，a source strength solving error of 0.8%，and a single computation time of less than 1 second. This enables fast and accurate positioning and estimation of source strength.

To address the complex scenarios of identifying danger zones in tower crane operations during construction，an early warning method of tower crane danger zone was proposed using computer vision technology. This method combined dynamic determination of tower crane danger zones with computer vision to detect personnel wearing situations of safety helmets and safety belt at the construction site and the inadvertent intrusion beneath the tower crane. Additionally，the YOLOv5 algorithm was adapted with attention models，and interactive window detection software was developed. Results indicate that the recognition accuracy of this model for human intrusion behavior and personal protective equipment exceeds 85%，demonstrating high precision. This method can be effectively applied in tower crane construction scenarios，optimizing fixed danger zone delineation to dynamic tower crane danger zones，and providing real-time monitoring of inadvertent personnel intrusion with warnings.

In order to improve the high-precision detection and early warning of crane safety operation and enhance the safety management ability of enterprises，focusing on the needs of unmanned industrial safety incident analysis and monitoring and early warning，an inspection robot that combines ground and air flight in hoisting scene was customized to intelligentize hoisting safety monitoring，pop-up image recording and safety alarm.A lifting dataset Cranes-Dataset (CRN-Dataset) containing 3 120 images was made，and DPIM algorithm was proposed to enhance the rapid detection ability of multi-scale objects.Based on corner detection and density-based spatial clustering of applications with noise and considering the safety attributes of the space distance between cranes and workers，the process of triggering alarms based on safety rules was developed to record real-time illegal operation image and popup alarm.The results show that，after actual deployment and verification，the DPIM algorithm significantly improves target identification ability compared with other traditional algorithms，and it is suitable for real-time calculation and data transmission of embedded edge intelligent analysis nodes to complete field deployment.

In order to address the challenges of urban traffic congestion and safety，an ACLR model was proposed. By integrating ConvLSTM，attention mechanisms，and residual structures，the ACLR model effectively enhanced the extraction of spatio-temporal traffic features.The time，space and other characteristics of taxi traffic were processed respectively，and the influence of regional point of interest(POI) data on taxi traffic was mined. Additionally，a specialized learning component was incorporated to capture the impact of external factors and point-of-interest density on traffic flow. Using taxi trajectory data from Beijing，the ACLR model demonstrates superior prediction accuracy compared to other models such as the autoregressive integrated moving average (ARIMA) model，long short-term memory (LSTM)，deep spatio-temporal residual networks (ST-ResNet)，convolutional neural network(CNN)-ResNet-LSTM (CRL)，and attentive crowd flow machines (ACFM) in urban traffic flow forecasting，which is helpful to improve the prediction performance of the model without POI density or considering POI density. The predicted value of the ACLA model is basically consistent with the real value，and it can also be in good agreement with the real value during peak hours，which effectively improves the ability to extract traffic temporal and spatial characteristics，reduces the prediction error，and optimizes the traffic flow prediction performance.

To solve the intelligent detection problem of fire lane occupancy warning，a lightweight early warning approach based on YOLOv7 was proposed by introducing the principles of area intrusion. Firstly，a research framework for detecting fire lane area intrusions was devised，utilizing the YOLOv7 model. This was accompanied by the compilation of an image dataset that encompassed fire lanes and vehicle detection，sourced from both field investigations and open datasets. Subsequently，the spatial pyramid pooling's multi-stage partial convolution was substituted with a receptive field block module，and the SimAM attention model was incorporated to enhance the network's capability in multi-scale feature extraction and fusion. Furthermore，the Slim-Neck architecture was implemented to minimize the model's computational requirements and parameter count. The interactive interface was then designed and implemented using PyQt5. The algorithm was subsequently validated in a community located in Xi'an，Shaanxi Province. The results show that the accuracy of the model to identify fire lanes and vehicles is over 80%. Compared with the original model，the improved model reduces the number of parameters by 20.5%，the floating-point calculation by 11.3%，and the detection speed by 42.4% to 48.6 f/s. This promotes the development of intelligent detection technology for fire lane occupancy.

In order to accelerate the speed and efficiency of autonomous systems testing，the method of generating a scene database for unmanned driving in campus environments was proposed. Firstly，the simulation test scenarios in complex campus environment were analyzed，and the campus scenes were simplified as a combination of road network structure，ground properties，interacting members and environmental factors. Secondly，the method of generating the scene database based on importance indicators was proposed to solve the boundedness of the campus scenario database. Then，the complexity indicators and interest probability indicators were used to describe the importance indicators of scenarios. The fuzzy analytic hierarchy process(FAHP) was used to evaluate the complexity of the scenario. The interest probability of the scenario was calculated by combining the kernel density estimation method and the interested weight calculation method. Next，the parameter space was segmented to obtain the set of similar scenarios，and the scenario sets were sorted according to test priority and importance indicators. The filtered scenarios were gradually added to the test scenario database，and the scenario database with test sequences was generated. Finally，the test evaluations based on the real-world campus scenario database were conducted to verify the effectiveness of the scenario database generation method proposed in this paper. The results show that the campus test scenes can be effectively described using four scene elements and the tree structure. The method proposed in this paper can generate a campus test scene library with high test efficiency，high coverage，conformity to natural probability，and interest interval，which is helpful to improve the efficiency of unmanned simulation test in complex campus environment.