Research On The Safety Evaluation Of CNG Filling Station Based On Bayesian Network

With the development of science and technology,energy conservation and the development of new energy vehicles become the mainstream of the times.As one of the new energy sources with less environmental pollution,natural gas has been widely used,and has led to the construction of a large number of compressed natural gas stations.However,natural gas is flammable and explosive,once triggered by accidents,the consequences are unthinkable.Therefore,it is very important to put forward effective measures to ensure the safety of gas filling station and control the accident inducing factors.Based on the Bayesian network theory and the characteristics of CNG filling station,this paper analyzes the risk factors that affect the safety of CNG filling station from the point of view of CNG station security,and constructs the CNG station safety evaluation index system with the help of GeNIe software.A Bayesian network model which can reflect the causality between factors is established,and the safety of Shandong Qufu CNG station is simulated and evaluated.The main contents are as follows:The main factors affecting the safety of CNG filling station are identified,and the safety evaluation index system is established.Fully considering many aspects that affect the safety of CNG filling station,such as environment,human behavior,management and equipment of gas filling station,etc..Author carries out research on all aspects,determines the specific characterization index factors contained in each aspect,and constructs the target layer,Criterion layer,index layer,three levels of CNG station safety evaluation index system.The safety evaluation model of CNG filling station based on Bayesian network is established.On the basis of constructing the index system,the security evaluation model of CNG station based on Bayesian network is established.Firstly,the network nodes and the index factors of each level of the index system are corresponding one by one,and the evaluation index system is transformed into Bayesian network structure.Then,the paper divides the security state level,designs the questionnaire to calculate the priori probability of each state level of the evidence node,and inputs it into the GeNIe software,sets up the parameter variable,creates the mapping,carries on the parameter study,updates the model.The conditional probability values of different states of residual nodes are calculated,and the security state of the research object is evaluated according to the principle of maximum membership degree.Finally,the evaluation results of Bayesian network model are verified and analyzed by reverse reasoning.The Bayesian network model is verified by take Shandong Qufu CNG filling station as an example.The simulation results of the model show that Shandong Qufu CNG station belongs to "medium",and the safety level is "3",that is,safety and risk are general.the influence of criterion layer factors on the safety of CNG filling station is ranked as facility equipment from large to small.The key factors affecting the safety of CNG filling station are purification system,staff quality,safety education,training and exercise,site selection and natural conditions.The verification results show that the safety evaluation results of CNG refueling station based on Bayesian network model are effective,which can not only provide the basis for the safety management decision of the filling station,but also avoid the risk for the gas station staff.It is of guiding significance to ensure the safe operation of the filling station.