To protect the occupational health and safety of ¹³¹I nuclear therapy workers and increase the utilization rate of ¹³¹I nuclear therapy，¹³¹I nuclear therapy workplaces in 10 hospitals in Shenzhen were selected in this study. The iodine box filter membrane sampling method，combined with a low background high-purity germanium γ energy spectrometer and passive efficiency calibration software，was used to measure the ¹³¹I activity concentration in critical areas of nuclear medicine workplaces. The accumulated effective dose of occupational personnel was estimated，and then the radiation safety risk of occupational personnel was evaluated. Furthermore，the effect of the correction factor change of ¹³¹I nuclide treatment operation on the radiation safety risk of occupational personnel was analyzed. The results showed that when the correction factor of ¹³¹I (liquid) nuclide treatment operation mode was set to 10，the radiation safety risk was far below the standard limit for manufacturer delivery and automatic packaging in nuclear medicine workplaces. It was recommended that the correction factor of the operation model of ¹³¹I (liquid) radionuclide therapy be distinguished by the packaging system，with manual packaging set to 1 and manufacturer delivery and automatic packaging set to 10.

In order to monitor the risk during the navigation of MASS，the safety control structure of MASS was constructed based on System-theoretic Accident Model and Process (STAMP). STPA was used to define the losses/accidents and system-level hazards，identify unsafe control actions，analyze loss scenarios，and construct an accident model for system state transition. FTPN was used to model the process model，and a given MASS navigation situation was used to obtain the relevant fuzzy time functions and to project the situational evolution of FTPN. A new risk level expression was introduced，and a two-dimensional path diagram of system loss/accident was used to visualize the real-time system risk level and system unsafe states transition paths. The results show that at the current moment of the set navigation situation，no safe water depth input，no updated collision avoidance path，unsafe heading and speed，and grounding are the highest risk system unsafe states and correspond to the four highest risk transition paths. The study shows that the FTPN process model driven by STPA can comprehensively assess the real-time risk level of MASS navigation. Visualize real-time risk with a two-dimensional path diagram of real-time losses/accidents of the system，which can monitor the unsafe system states during MASS navigation and describe their transition paths.

To solve the problems such as forward tilt deformation，serpentine shape，axis deviation and correction during shield tunneling，which affected the safety and efficiency of shield construction，a multi-objective optimization method of shield attitude combining CatBoost and NSGA-Ⅲ was proposed. Taking Guiyang Metro as the background，22 influencing factors were selected as input parameters，and the nonlinear mapping function relationship between input parameters and shield attitude was established by using CatBoost algorithm. The importance of input parameters was evaluated by random forest (RF) algorithm. A CatBoost-NSGA-Ⅲ multi-objective optimization model was established to minimize the absolute value of the shield attitude，and the applicability and effectiveness of the proposed method were verified by a case study. The results show that the prediction model obtained by using CatBoost algorithm to train engineering measured data has high accuracy，and the R² range of 5 shield attitude targets is 0.916-0.943. By using the CatBoost-NSGA-Ⅲ multi-objective optimization method，the attitude of the shield can be optimized significantly，and the average value of the overall improvement is 53.34%.

Debris flow，as a common geological disaster，has a complex formation mechanism with numerous influencing factors and multiple uncertainties. To comprehensively consider the synergistic effects of various influencing factors，based on information fusion and uncertainty analysis theory，this paper proposed a debris flow susceptibility evaluation method based on evidence theory and cloud model. Firstly，the BPA function of key evaluation indicators for debris flow susceptibility was calculated using a connection cloud model. Subsequently，the reliability and uncertainty of the indicators' BPA were modified using Lance distance and DENG entropy，respectively，resulting in a corrected BPA. Finally，evidence fusion was performed on the corrected BPA based on Dempster-Shafer (D-S) evidence theory to achieve debris flow susceptibility assessment，followed by a case validation. The results show that the connection cloud model used in this paper overcomes the limitation that the normal cloud model requires indicators to follow the normal distribution when calculating BPA，and it considers the randomness and uncertainty of indicator distribution. The proposed method's evaluation results are generally consistent with those of four other commonly used evidence fusion methods，proving it to be effective and feasible for debris flow susceptibility evaluation. The conflict evidence fusion method improved based on Lance distance and DENG entropy can enhance the convergence speed and precision of evidence fusion，making the results more accurate and reliable.

An open-pit mine landslide identification method was proposed based on object-oriented annotation datasets and the Res-U-Net model to realize accurate identification and early warning of open-pit mile landslide disasters. Firstly，the mine landslide image data in the study area were obtained by UAV aerial survey. Secondly，the multi-scale-spectral segmentation method and threshold separation principle were applied to divide and classify the open-pit mine landslide data，and the landslide dataset was developed based on the object-oriented method. Then，the U-Net network was used as the infrastructure to propose a landslide identification semantic segmentation model based on Res-U-Net by integrating the residual module into each convolutional layer. Finally，the datasets constructed by different methods were used to identify landslides，and the Res-U-Net model was compared with the widely used semantic segmentation models，Fully Convolutional Networks (FCN)，and U-net. The results indicated that the landslide data set based on object-oriented annotation had better landslide identification performance when compared to the traditional manual annotation dataset，resulting in improvements in identification accuracy，recall rate，F₁ score，and kappa coefficient of more than 12%. The landslide identification accuracy of the Res-U-Net model was more than 0.8，realizing the accurate landslide open-pit mine disaster identification.

In order to mitigate the impact of traffic accidents on the operation efficiency of freeways，and improve the throughput capacity of the accident area，based on real-time information via vehicle-to-vehicle and vehicle-to-road，a cooperative lane change guidance strategy in freeway traffic accident areas was proposed with the safety potential field theory. Firstly，in view of various traffic accidents in different lanes in two-lane or multi-lane traffic of one-way，the guidance area of collaborative lane change was divided into four areas: accident protection area，guidance transition area，collaborative lane change guidance area and free lane change area. The guidance threshold of lane change was determined to update vehicle status. Furthermore，the safety potential field of traffic accidents was established，and the corresponding guidance strategies of vehicle cooperative lane change were proposed according to different scenarios. The calculation methods of the safe distance for vehicle lane change and the latest lane change position were given. Finally，based on simulation of urban mobility (SUMO) software，the simulation results were verified in various scenarios. The results show that in the two-lane of one-way scenario，the optimization effect is most obvious when the vehicle cooperative guidance rate is at 75%. In the multi-lane of one-way scenario，the optimization effect is most obvious when the guidance rate is at 50%. Meanwhile，through comparative analysis，it is found that the average speed of vehicles passing through the accident section is increased by up to 6.3% and the maximum vehicle delay is reduced by 14.6% after adopting the lane change guidance strategy.

In order to solve the problems of difficulty in quantifying the risk of airport runway incursion events，poor timeliness and low accuracy，and to enhance the capability of predicting runway incursion risks，a DBN model incorporating reinforcement learning for risk prediction was constructed. Firstly，causal inference theory was combined with grey relational analysis to analyze historical runway incursion events and identify the underlying risk factors. Secondly，Bayesian network(BN) theory was applied to explore the correlations among these factors and quantify these correlations using the Pearson linear correlation coefficient. This process helped in constructing a causation correlations network that effectively represented the propagation of risks associated with runway incursions. Then，the triangular fuzzy method and Hidden Markov Models (HMMs) were utilized to further refine and optimize the DBN parameter learning mechanism. Finally，the model's accuracy was validated using historical data. The results demonstrate that the proposed model's predictions of runway incursion risks closely align with the statistical values of historical data，achieving an accuracy rate of 84%，which represents a significant 10% improvement over Bayesian network predictions. Additionally，the use of mutual information to identify key nodes is found to effectively improve accuracy and discrimination compared to the degree value evaluation method.

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.

In order to clarify the research progress of intelligent risk management in coal mines，the research status of data-driven coal mine safety risk management models was comprehensively analyzed. The prediction methods and analysis models for coal mine safety risk assessment were also reviewed. Firstly，the intelligent risk management was defined，and the scope of analysis was determined by searching relevant literature. Then，the research status，existing problems and development trend of accident big data were reviewed from three aspects: data-driven analysis method，coal mine safety risk assessment model and coal mine big data prediction and early warning platform. The results show that the theory and application framework of data-driven risk analysis in the field of coal mine safety has been basically formed，but it still cannot meet the needs of risk assessment and emergency management. In the application of early warning platform，a unified and general basic framework of big data analysis platform for coal mine safety production has been formed，but its application and promotion in production practice are far from enough. In the future，it is necessary to construct the comprehensive risk assessment model to study the risk of coal mining，starting from improving data quality and integrating dynamic and static multi-source data. Besides，the application of data-driven analysis in production practice should also be strengthened. These works can promote the transformation of coal mine safety risk management mode from empiricism to data-driven，and realize the informatization and intelligence of coal mine safety risk management.

In order to monitor the health state of construction formwork support systems and prevent the risk of safety accidents caused by formwork collapses，a new intelligent monitoring method combining EMI and CNN for joints of formwork support systems was proposed. Firstly，based on the electromechanical coupling and sensing-driving characteristics of PZT，PZT-joint coupling model was built based on the electromechanical impedance sensing mechanism. Secondly，the original conductivity of PZT patch，coupled with the monitored structure，was used as a monitoring signature for identifying joint looseness based on the EMI technique. Thirdly，EMI-CNN model was built with the 801 original conductance signals of PZT over the sensitive frequency range as the inputs，and the nine degrees of joint looseness as the outputs. In total，the dataset consisted of 189 samples，162 for training and 27 for testing. At last，taking an actual formwork support system joint from building site as an example，EMI-CNN model was verified and compared with EMI-BP model by the experiment. The research results show that EMI-CNN model reached convergence after 85 iterations. The prediction accuracy of the EMI-CNN model reached 100%，which is 29.63% better than EMI-BP model. This proposed method is distinguished by its real-time，accurate and non-destructive monitoring capabilities，providing an effective solution for health monitoring of joints in construction formwork support systems.