In order to reduce the risk of traffic accidents in expressway agglomerate fog environment, the expressway agglomerate fog zone was divided into transition zone and conflict zone, and driving simulation tests were designed and carried out. Firstly, Analysis of Variance(ANOVA)and non-parametric tests were used to analyze the driving behavior in the two zones. Then, Pearson correlation analysis was used to explore the correlation characteristics of the driving behavior indicators between transition zone and conflict zone. Finally, a binary logistic regression model was used to explore the mechanism of the influence of driving behavior on the collision risk. The study shows that: in transition zone, driving experience and visibility change rate have a significant effect on speed standard deviation and inner boundary speed. Compared with novice drivers, experienced drivers have 36.8% lower speed standard deviation and 13.5% higher inner boundary speed. Visibility change rate has a statistically significant difference on speed standard deviation and inner boundary speed. Different visibility change rates correspond to different inner boundary speed and speed standard deviation. In conflict zone, time to collision (TTC) was significantly positively correlated with visibility, and reaction time was significantly negatively correlated with visibility. TTC, reaction time and collision rate were all significantly negatively correlated with the conflict distance. The speed adjustment behavior in the transition zone significantly affects collision risk in conflict zone.

In order to solve the problem of ambiguity and limitation of application scope in the establishment of safety prevention and control measures for high-temperature tunnel construction, a method of generating prevention and control measures for high-temperature tunnels based on C-CBR is proposed. Firstly, the index system of the cloud case base is constructed for the tunnel high ground temperature and heat damage problem and combined with the analytic hierarchy process(AHP)-entropy weighting method for the comprehensive weighting calculation. Secondly, the cloud historical case base is constructed based on the three basic characteristic indexes of the cloud model, namely expectation, entropy, and hyperentropy, and the principle of case-based reasoning, and the collaborative filtering algorithm is applied to preliminarily filter the historical cases which are in line with the requirements. And then, the cloud model algorithm which integrates the shape-distance is introduced to achieve the integration between target cases and historical case base. Then, the integrated shape-distance cloud model algorithm is introduced to achieve similarity matching between the target cases and the historical case base, to correct the similarity and update the historical case base to generate the optimal similarity prevention and control measures; finally, the optimal similarity prevention and control measures are fully compared with the actual prevention and control measures in the field. The results show that the corrected similarity between the optimal historical case and the target case generated based on C-CBR is 0.746 0, which is a good match, and the the target case similar prevention and control measures are basically in line with the actual prevention and control measures of the project, which makes the generation method more applicable to different tunnels with high ground temperatures under construction.

In order to investigate the variation characteristics of flue-gas temperature at different locations of flame retardant belt during a fire, a fire test was conducted on flame-retardant conveyor belts in a full-scale roadway. First, six temperature measurement sections were arranged along the axial direction in the roadway, and five thermocouples were installed at each section to record the flue-gas temperature at different locations during the combustion process. Then, a fire characteristic test was conducted on a 4 m × 1.2 m (length × width) steel wire rope core flame retardant belt under the condition of wind speed of 1.2 m/s in the full-scale roadway. The experimental results show that the distance from the fire source determines the flue-gas temperature and temperature rise process in the downstream area of the fire source. The closer to the fire source, the stronger the heat effect caused by the combustion of flame retardant tape, and the higher the flue-gas temperature at the corresponding location. Since the tape fire is an unsteady combustion process, and the change in smoke temperature at different sections after a period of time is nonlinear related to the distance from the fire source. In a horizontal roadway at the same height, the flue-gas temperature on both sides of the roadway is higher than that in the middle of the roadway. As the distance from the fire source increases, the temperature difference between the middle and the sides of the roadway gradually decreases. In the longitudinal range of the roadway downstream of the fire source, the flue-gas exhibits a stratified state, and the flue-gas temperature decreases gradually from the roof to the floor of the roadway. There is an obvious temperature difference between flue-gas layers at different heights, and the temperature difference between the flue-gas layers decreases gradually with increasing distance from the fire source.

In order to study the SFEP and its characteristics, a method for describing SFEP using quantum concepts is proposed. The relationship between SFEP and event function states was discussed. The principle of describing SFEP by quantum thought was studied. The mathematical model of quantum description of SFEP was constructed. The measurement collapse of SFEP quantum states was realized, and the failure mode and its probability were finally determined. The results indicate that SFEP has diversity and uncertainty, event quantum state superposition, measurement collapse, bipolar state, which is the basis of studying SFEP through quantum thought. SFEP can be decomposed into multiple layers, with distinct objects and quantum state superposition modes in each layer. The SFEP quantum states are derived from the superposition of the quantum states of objects at each layer, which is composed of the polynomial product of the probability amplitude of the objects in each layer. The restriction conditions for the formation of the probability amplitude of quantum states of each layer are given. Each dimension of the SFEP quantum state vector is a fault state. The measurement of the factor phase value makes the SFEP quantum state collapse, and the possible failure mode and occurrence probability are obtained.

In order to constrain the definition and scope of marine nuclear safety and to ultimately support the comprehensive objectives of protecting human health, guaranteeing radiation safety for the marine environment and organisms, and maintaining marine biodiversity and ecological service functions, the evolving concepts from safety, marine safety, and nuclear safety, to marine nuclear safety were comprehensively tracked. Assessing techniques of marine nuclear safety including background baseline method, activity limit method, dose limit method, and radiation risk method were established to evaluate the discharging of artificial radionuclides into the ocean. To protect human health and ecological and environmental safety, priority directions including source identification, bioconcentration, and transfer of radionuclides and their associated nuclear safety assessment in seafood based on "Big Food Concept", assessment of radiation dose on marine biotas and protection, and radionuclides-driven synergistic effects of multiple pollutants were proposed and suggested for the future development of marine nuclear safety. In summary, this study systematically examines the conceptual framework and scope of marine nuclear safety, evaluates the advantages and applicability of four assessment methodologies to address multi-level nuclear safety management needs, and identifies three key research directions to advance the high-quality development of marine nuclear safety assessment technologies in the context of climate change and human activity.

In addressing the issues of non-compliance by ground service agents and lax supervision by airlines in the transportation of hazardous materials in civil aviation. Firstly, evolutionary game models between airlines and ground service agents were constructed under both static and dynamic punishment mechanisms. The evolutionary stable strategies under different mechanisms were then explored. Subsequently, a quantitative analysis model was developed by integrating system dynamics to further analyze the interactions between the two parties. Finally, simulations were conducted to analyze the impact of key parameters on the behavioral strategies of both sides. The results reveal that under the static punishment mechanism, no stable equilibrium point is observed in the game system. No evolutionary stable strategy is formed, with the behavioral strategies of both parties showing periodic fluctuations over time. In contrast, when a dynamic punishment mechanism is introduced, a stable equilibrium point emerged in the game system. The behavioral strategies of both parties converge to a stable focal point. Moreover, compared with a low level of punishment, a higher level of punishment is found to be more effective in increasing the probability that ground service agents strictly comply with the agency agreement.

In order to solve the problem of difficulty in determining the number of modes and quadratic penalty factors in the process of variational mode decomposition, a GWO-VMD algorithm was proposed. Based on the Longlong Tunnel gas explosion method construction, an integrated acquisition module was used to collect vibration signals during the rock-breaking process. The phase space reconstruction recursive graph (RP) similarity model was used to accurately distinguish the GWO-VMD principal components of the signals. The real signal with interference removed was reconstructed, revealing the distribution characteristics of the energy of the gas explosion signal in the time-frequency domain, and quantifying the accuracy error of the digital electronic detonator. The results show that compared with traditional variational mode algorithms, the GWO-VMD algorithm has significant advantages in improving the signal-to-noise ratio and adaptive feature extraction of gas explosion signals, and has strong time-varying frequency tracking performance. It can accurately identify the detonation accuracy of digital detonators and effectively identify the related of tunnel-blasting detonator disaster sources.

In order to explore the interaction mechanism among various behaviors in coal mine safety management, firstly, the coal mine safety management behaviors were divided into organizational safety behavior and individual unsafe behavior. Then, four dimensions and 16 sub-dimensions of organizational safety behaviors were identified, and 11 types of unsafe behavior were also identified. Finally, the interrelationships of different dimensions of organizational safety behaviors were explored, and the interaction model of coal mine safety management behavior was constructed. The results indicate that the highest degree of cause is safety supervision and inspection, and the lowest degree of cause is safety accident management. Safety education and training plays an important role as an intermediary and a bridge. The higher centrality of implementing training programs and providing adequate and effective safety training indicates that they dominate the interactions of organizational safety behaviors. The most common unsafe behavior is the failure to effectively stop and correct violations. The interaction model of coal mine safety management behavior includes a driving mechanism and a feedback mechanism.

In order to improve the work safety management level of China's chemical (hazardous chemicals) industry, firstly, the formation of chemical process safety management was expounded, including the traditional management method and the proposal of process safety concept. Then, the characteristics and specific content of the elements of the chemical process management elements system in China were analyzed, including 8 modules and 20 elements. Finally, a systematic solution was proposed, that was, four-element solution with safety leadership, management barrier (responsibility, equipment, management and capability), and safety culture as the primary-level elements. The results show that core elements of four-element solution are the concept, safety leadership, management barrier (responsibility, equipment, management, capability) and safety culture. Among them, safety leadership is the most critical element, leading the concept into practice, responsibility division, equipment upgrading, management system construction, capability improvement and culture cultivation. Other elements together constitute the skeleton of work safety management, and realize systematization. Businesses can help prevent accidents and improve the overall safety level of the industry by taking the lead and promoting the implementation of various elements through participation of all staff.

In order to explore the effect of gas injection pressure of power plant on the adsorption gas desorption efficiency of coal seam, NMR technology was used to quantitatively study the dynamic change process of CH₄ competitive adsorption under different gas injection pressure of power plant. The influence law of gas injection pressure on methane desorption, desorption efficiency and unit time recovery efficiency was obtained. The experimental results show that when anthracite coal adsorbs CH₄, the T₂ spectrum shows a bimodal characteristic. As time increases, the peak of T₂ spectrum of adsorbed CH₄ increases, while the peak of T₂ spectrum of free CH₄ decreases. With the increase of time, the adsorption rate of CH₄ by coal slows down. During the competitive adsorption process of power plant flue gas and CH₄ from coal, the T₂ spectrum exhibits three-peaks characteristics. As time progresses, the T₂ spectrum peaks of adsorbed CH₄ decreases, while the T₂ spectrum peaks of free CH₄ increases. The integrals of adsorbed and free T₂ spectra show an exponential function with the adsorption time, and a linear function with the pressure of flue gas injection in the power plant. The power plant flue gas injection pressure shows a linear increase with CH₄ desorption efficiency. The injection of 3.5 MPa power plant flue gas increases the total desorption of adsorbed CH₄ from coal samples in the sample tank by 0.016 mol, enhances the recovery efficiency per unit time by 3.440%, and increases the total desorption efficiency by 20.631%, compared to 1.5 MPa.