To investigate the source and proportion of thermal runaway heat transfer of lithium-ion batteries in confined conditions space such as aviation transportation，the 18650 lithium-ion battery (100% state of charge) with lithium cobalt oxide (LCO) as the cathode material was used. The heat transfer model of thermal runaway of lithium-ion batteries was proposed by ANSYS Fluent software. Furthermore，the pyrolysis gas generated by the first battery and its thermal runaway was used as the heat source，and the second battery was heated to thermal runaway through radiation and convection. The results showed that when 2^nd battery reached the thermal runaway temperature，the heat generated by the internal side reaction accounted for 30.01% of the total energy. The gas combustion generated by the 1^st battery thermal runaway provided energy for 2^nd battery thermal runaway，accounting for 5.64% of the total energy. When 2^nd battery reached the maximum temperature，the heat generated inside the battery accounted for 87.39%，and the energy provided by the gas combustion was 1.76%. the pyrolysis gas combustion accelerates 2^nd battery's thermal runaway，though it is a heat source，it is not a heat source. Although the combustion of pyrolysis gas accelerated the thermal runaway process of 2^nd battery，the proportion of energy provided was relatively small.

In order to explore the mechanism of rock burst induced by the instability of coal rock combinations under different dip angles，a particle discrete element program was used to conduct uniaxial compression tests on five sets of coal and rock combination models with different dip angles: 0，15，30，45 and 60°. The research results indicate that coal is the main cause of instability and failure in coal-rock combinations. When the dip angle of the coal-rock combination increased from 0° to 30° and from 30° to 60°，the uniaxial compressive strength of the coal-rock combination decreased by 2.01% and 9.59%，and the number of microscopic cracks decreased by 22.9% and 4.0%，respectively. The appearance time of the acoustic emission signal is advanced，indicating that the increase in dip angle led to a decrease in the uniaxial compressive strength of the coal-rock combination，and the instability failure time is advanced，but the degree of failure is reduced. In the early stage of uniaxial loading，the movement of coal and rock particles at the interface of different dip angle combinations led to an expansion trend at the interface. The dip angle affected the movement of coal and rock particles near the interface，leading to a gradual transition of the failure area of the combination from the coal body to the coal-rock interface. When the cracks in the coal body extend to the coal-rock interface，the interface slip effect generated by the high dip angle coal-rock combination causes its failure mode to change from compression shear failure to slip failure，with a 30° dip angle as the boundary.

In order to investigate the influence of porous materials on the explosion propagation characteristics of LPG in vent pipes，experiments were conducted on a self-built gas explosion test platform. The inhibitory effect of silicon carbide materials with different thicknesses and porosities on the flame propagation characteristics and explosion overpressure in the vent pipe was investigated. The results show that porous materials can effectively prevent the propagation of LPG explosion flame. With the increase of porosity and thickness，the propagation distance，propagation speed and intensity of LPG flame are reduced，and the porosity has a greater effect on the fire resistance than the thickness. The installation of porous materials also reduces the maximum explosion overpressure. When LPG is not ignited behind the installation location，increasing the porosity and thickness of porous materials，the maximum explosion overpressure in front of the material (PT1，PT2) increases，and that behind the material (PT3) is close to the static film-breaking pressure. The firestopping effect of material porosity and thickness shows that increasing the porosity of the material for firestopping should be prioritized when the strength of the material is up to standard in engineering applications.

In order to explore the influence of different learning scenarios on the learning effect of learners in the teaching process of fire emergency fire escape in subway stations，and explore the advantages of virtual reality learning. Questionnaires and MP160 physiological instrument were used to obtain different behavioral response data of learners in paper learning，mobile video learning and virtual reality learning scenarios，and these data were processed and analyzed. The results show that there are significant differences in learners' academic performance under the three learning scenarios. Compared with mobile video learning and paper learning scenarios，virtual reality learning scenarios can promote learners' understanding of subway fire escape knowledge and significantly improve academic performance. Under the three learning scenarios，there are significant differences in learners ' interest in learning. Virtual reality learning scenarios show significant advantages in improving learners ' interest in learning subway fire escape knowledge. The virtual reality learning scene reflects a higher level of physiological arousal in terms of physiologic index.

In order to prevent the risk of uncertainty and extreme impact brought by the black swan，the origin，transmutation and prospect of the black swan were studied in depth under the perspective of metaphor theory. Firstly，the conceptual integration theory was adopted to analyze the cognitive construction mechanism of the black swan risk metaphor from the aspects of historical flow and development trend，summarize the overall development trend of the black swan，and on the basis of this，divide the black swan into three important transmutation stages of founding，exploring development，and innovating and perfecting，so as to put forward the research dilemma and future outlook of the black swan under the background of digital intelligence empowerment. The results show that the risk metaphor meaning of black swan is a risk event with rarity，episodic，unexpected and unpredictable，and extremely negative impact; the development of the black swan risk metaphor will enter the stage of theoretical differentiation of quantitative fluctuation and change，and the direction of the research should be from generalized application to precise application，from single risk to composite risk，and from silo research to group research.

In order to solve the nonlinear problem of dynamic loss of soil around the pipe during the collapse development process of the buried pipeline，first，a numerical analysis method was used to construct a nonlinear coupling model of pipe-soil that passes through the collapse zone. Then，the model was verified based on experimental measured data and specifications. Finally，a study on the damage mechanism of buried pipelines subjected to soil collapse was carried out，and the dynamic evolution process and mechanical characteristics of buried pipeline collapse were discussed. The results show that the axial stress is the control stress，the mid-span pipe bottom is the control point，the mid-span section is the dangerous section，and excessive tensile stress is the main reason for failure of buried pipelines. It is confirmed that "pipe-soil separation" phenomenon exists. When the collapse depth reaches 48 mm，"pipe and soil separate"，as the collapse process progresses，the entire process of pipe-soil structure from the beginning of deformation to the tensile failure of buried pipeline can be divided into three stages: the top pressure stage，the transition stage and in the bottom tension stage，the collapse depth of 80 and 160 mm is the dividing point. When the collapse depth reaches 59 mm，the pipe top stress changes from valley to peak at mid-span position. When collapse depth reaches 80 mm，friction stress appears.

In order to solve the problem that the pressure drop signals caused by compressor suction or upstream block valve cut-off conditions leaded to incorrect shut-off of the block valve，and the problem that the block valve failure due to insignificant pipeline pressure drop caused by small hole leakage，a simulation model was established. Taking a typical gas transmission trunk line as the research object，300 sets of pressure drop signals under three different working conditions，namely compressor suction，emergency cut-off of the block valve and pipeline leakage，were obtained. The pressure drop rate of the pressure drop signal was calculated by point-to-point detection method. Singular value decomposition(SVD) method was used to extract the characteristics of the pressure drop rate signal，and the min-max normalization method was used to normalize the characteristic values of the pressure drop rate signal. SVM method was used to identify the characteristic value signals of different pressure drop rates，and the corresponding working conditions were obtained. To solve the problem that the unreasonable setting of kernel function parameters and penalty factors in the SVM model affected the accuracy of algorithm recognition，TLBO algorithm was used to optimize the kernel function parameters and penalty factors，and a TLBO-SVM model for intelligent identification of gas pipeline leakage signals was established. The model was applied to classify and identify 300 groups of simulated pressure drop rate signals in three working conditions. The results show that the recognition accuracy of the model is 92.22% for three kinds of pressure drop rate signals under different working conditions. The identification accuracy is 96.67% for small hole leakage with a leakage diameter of 50-125 mm and a pressure drop rate range of 0.01-0.07 MPa/min. For the actual leakage pressure drop rate signal of a main pipeline，the accuracy of TLBO-SVM is 100%.

In order to improve the performance of small target detection in infrared imaging and the ability of low altitude airspace supervision，an infrared small target detection network based on multi-scale attention feature enhancement fusion was proposed. Firstly，Resnet34 was used to extract the multi-scale features of infrared images. Secondly，the multi-scale spatial attention feature enhancement module(MFEM) was used to improve the ability of feature extraction. Then，in the step-by-step up sampling process，the dual channel attention feature fusion module(DFFM) was used to fuse the semantic information and detail information to better protect the characteristics of infrared small targets. Finally，taking the video sequence detection of ground/air infrared dim small aircraft target as an example，the real scene test was carried out by comparing with other methods. The results show that compared with existing methods，the proposed method improves the scores of intersection over union(IoU)，F-measure and false negative rate(FNR)，and can accurately locate the target and generate good segmentation results. The DFFM can simultaneously use multi-scale context information and spatial attention mechanism to highlight infiared small targets. The DFFM assigns weights to sets of different channel features，thereby obtaining the most appropriate feature map for feature fusion and improving the detection performance.

In order to improve air traffic controllers' emergency response ability，a reliability model of air traffic controllers' SA was constructed from three aspects: air traffic controllers' personal ability，air traffic control task characteristics and air traffic control equipment. Tower control simulation software and SA Global Assessment Technique (SAGAT) were used to measure the air traffic controllers' SA level in emergency situations. Based on BN，SA reliability was quantitatively analyzed to predict air traffic controllers' level of SA，and based on Bayesian inference，the key factors affecting the reliability of air traffic controllers' SA were analyzed. The results show that the reliability of SA of air traffic controllers is positively correlated with the level of SA in emergency scenarios，and the SA level can be predicted by air traffic controllers' SA reliability. The BN inference analysis reveals that the factors with a higher degree of influence and sensitivity to the reliability of controllers' SA were the deployment of aircraft availability time，the accuracy of the control equipment and the controller's memory. The causal chain that has the greatest impact on the air traffic controllers' SA reliability is the deployment of aircraft availability time→mission characteristics→SA reliability.

Frame-rocking wall was a composite self-resetting structure that could effectively improve the seismic resistance and toughness of buildings. To fully understand the random response characteristics of structures under earthquakes，a simplified nonlinear equation for a multi-degree-of-freedom frame-rocking wall structure was constructed，and an equivalent linear dynamic equation with time-varying parameters was constructed based on the assumption of pseudo harmonic behavior in response using equivalent linearization. Further，based on the principle of random averaging，the Fokker-Planck-Kolmogorov(FPK) equation could be derived to determine the time evolution form of probability density function(PDF) for controlling the amplitude of the response，and ultimately the first-order differential equation for the time-dependent variance of the random response could be obtained. Finally，a computational model was constructed using a framework of a certain teaching building as a sample for validation. The results show that the approximate analytical method has excellent accuracy，and while ensuring the accuracy of the random response time-related variance，it can improve the efficiency of analysis compared to traditional Monte Carlo simulation(MCS) methods. In the results of non-steady ground motion power spectrum models in separable and non-separable forms，the trend of the random response variance curve is related to the form of random seismic excitation，and its segmentation points show obvious unsmooth phenomena under the action of segmented modulation of non-stationary spectra. The results under different types of random seismic excitation disturbances demonstrate the excellent applicability of this method.