Study On The Safety Of Imported Dangerous Goods Such As Oil And Gas Energy

In today’s petrochemical production,oil and gas explosion is the main cause of major disasters.Flash point(FP)is the basis for classifying flammable liquids and their classification criteria,and also the key parameter for measuring fire hazard.Explosion limit is an important risk parameter for evaluating the fire risk of combustible gas and formulating relevant standards and norms.Based on the components of liquefied petroleum gas(LPG)and liquefied natural gas(LNG)in Beihai,Guangxi,this paper simulates LNG and LPG,and systematically studies the factors affecting the explosion limit by using the European standard "bomb method" explosion instrument,and predicts the flash point by using the Engler range temperature model of foreign oil products.The main results are as follows:First,the Le-Chatelier formula is used to calculate the explosion limit of mixed gases at room temperature and pressure.Based on this formula,the explosion limit of LPG and LNG is measured by the standard BS-EN-1839-2003-2012.The influence of initial temperature on the explosion limit is measured.With the increase of temperature,the upper explosion limit decreases,the lower explosion limit widens,the upper explosion limit and the explosion limit widen.The lower bounds are highly linear with temperature.Nitrogen is more sensitive to the upper explosion limit,while the lower explosion limit is almost unchanged.For LPG with different composition and content,the explosion limit also changes accordingly.When the content of components changes,the trend of explosion limit tends to increase with volume fraction.The method of judging explosion pressure rise is more reliable than the method of visual flame propagation distance.In this work,based on the flash point and EN’s distillation temperature of 61 foreign crude oils,the flash point of crude oils was predicted by using three models,MLR,BP-ANN and RBF-ANN,respectively.The results show that the predicted and experimental values are in good agreement.The validation of multivariate linear regression model shows that there is a certain non-linear relationship between the temperature and flash point of Engler distillation.BP-ANN model is slightly better in terms of neural network method.