In view of the shortcomings of the existing accident causation models, such as simple listing and superposition of causative factors, not considering the coupling effect between factors, unable to make quantitative analysis and so on,a new nonlinear multi⁃factor coupling accident causation model was proposed and applied to the prevention of oil depot accidents. In the new model, the calculation formulas of the main influencing factors, the influence coefficient and the multi factor coupling rheological nonlinear damage were proposed. The most prominent feature of the model is that it can distinguish the dynamic influence of the rheology of the main influencing factors on the damage of the complex system under the coupling action of multiple factors. MATLAB software was used to simulate the changing state of the oil depot system after single factor rheology, main influencing factor rheology and non⁃main influencing factor rheology. The results show that the main factor affecting the safety of oil depot is the lack of safety consciousness,and the lack of safety consciousness of oil depot personnel has the greatest influence on other factors. For oil depot accident prevention, according to the nonlinear multi⁃factor coupling accident cause model proposed in this paper, it is necessary to take effective measures to control the main influencing factors and seize the main time nodes.

Based on the traditional prediction model of triangular fuzzy accident tree, which had the defect of large error caused by linear approximation, the source of error was analyzed according to the truncated set theorem, and the improved algorithm for drawing membership curve of triangular fuzzy number of accident treetop events was given according to the representation theorem. Matlab software was used to program, the modified and traditional triangle fuzzy accident prediction algorithms were used to predict the electrostatic accumulation accidents in oil depots, and calculated the fuzzy importance of 8 causes. According to the calculation results, compared with the traditional algorithm, the average error value of the fuzzy importance calculation of the eight basic events is reduced by 11.91%，which shows that the modified algorithm greatly optimizes the shortcomings of the traditional algorithm.

In order to solve the problem of complex system in the security field by theory and computer， the objectives, conditions, and characteristics of incompatible problems in the field of security were defined from the perspective of acceptable security state value and the four elements of security on the basis of studying the theory and method of inconsistent problems, and the corresponding mathematical model was given. The key conjugate elements for the extension of the incompatible problem were defined: The hard part was the safe person and the safe thing, and the soft part was the safe society and the security system. The mathematics strategy to solve the incompatible problems in a gas station, the result showed that the application of this method could solve the problem of the safety of the gas station and provide the new ideas and the basis for safety performance management.

Once hydrogen sulfide is leaked and diffused, the consequences are unimaginable. In order to master the law of hydrogen sulfide diffusion, the diffusion law of hydrogen sulfide under the influence of source strength, wind speed and source height was numerical simulated by MATLAB and VC++ software. The simulation results show that at night, when the source strength increases, the downwind direction diffusion range increases under the leakage source, the hydrogen sulfide mass concentration gradient increases in the diffusion area, and when the source height increases, the maximum width of the downwind direction diffusion area under the leakage source becomes narrower and the area becomes smaller. During the daytime, when the sunlight intensity is low and the wind speed is greater than or equal to 4 m/s, the downwind direction diffusion range of the leakage source decreases as the wind speed increases, and the hydrogen sulfide mass concentration gradient decreases in the diffusion area. Increasing the source height and wind speed is beneficial to reducing the mass concentration of hydrogen sulfide and the hazard. The research results can provide a basis for accident prevention and emergency rescue.

The DEMATEL algorithm is an algorithm for analyzing the relationship between the influencing factors of complex systems. The algorithm has important application value for identifying the key factors of complex security system. First, on the basis of the DEMATEL algorithm, according to the knowledge of cognitive psychology and the status quo of the safety management of a gas station, the complex gas station safety system is divided into 4 subsystems, and further refined into 31 specific factors. Secondly, according to the DEMATEL algorithm, the MATLAB program is compiled to divide the cause factors and the result factors, and the influence degree, cause degree and center degree of each factor are calculated. Finally, the location of the key factors are displayed intuitively by drawing the cause results diagram. The results show that the information channel interruption, lack of recognition and information uncertainty, contingency plans and response defects are the key factors affecting the safety of gas stations. The longterm solution for gas station safety management is to control these 4 factors.

Because there is a large error defect of predicting the severe fluctuating fire data by traditional grey forecasting model, the article expends the original data sequence of grey prediction model using Taylor formula and expends Lagrange type remainder, which modified the traditional grey forecasting model and improved the traditional grey prediction algorithm. The article used Matlab software to program, and used the traditional gray prediction algorithm and improved gray prediction algorithm,respectively.Then the article selected three sets of data to predict aiming at the mini zone bathtub shape fire accident. As the results show that: modified grey forecasting model prediction curve gives satisfactory accuracy comparing to the characteristic of bathtub curve, which has a high consistency with the original data.And the poor quality, large error and low accuracy is improved. Compared with the traditional grey prediction model, the error of the improved grey prediction model has reduced 86.59%, 55.32%, and 55.70%, respectively. The prediction results of the improved grey prediction model can meet the requirements in three aspects: accuracy, relative error and correlation.

The probability of a number of basic events itself was uncertain and fuzzy when the traditional FTA method was used in the practical quantitative analysis quite. It was difficult to obtain the exact probability from failures, which violated the premise of exact probability selected. In order to solve the traditional fault tree analysis (FTA) of the probability the basic event fuzzy and uncertain problems, the concept of fuzzy theory were introduced, and the fuzzy probability was used to deal with the basic events lack of statistical data. The basic events of fuzzy probability were described by using a clear triangular fuzzy number, the fuzzy probability distribution of top event layer by layer was obtained, and the fuzzy importance of each basic event was determined by using fuzzy numbers value method. And the sort of top event of failure factors was obtained, the main factors influencing the accident were identified and effective measures from fundamentally preventive and the key points to control the top event were proposed.

According to the process characteristics of flashing retorting process, a set of suitable for high temperature oil washing oil vapor recovery technology was designed, combined with pilot plant field data, using plus Aspen software for process simulation. Shale oil was separated to heavy oil, heavy diesel oil, light oil fraction and gasoline fraction by the oil washing method. The process produced steam as byproduct and it had the advantages of high recovery rate, low oil loss, low energy consumption, and could effectively avoided environmental pollution. Using this process, the rate of gas recovery could be greatly enhanced: crude oil washing tower could recycle 39.7% of the oil steam, heavy diesel oil washing tower could recycle 60.3% of the oil steam. Compared with the traditional process, it could improve the recovery rate of 10%-20%.