| At present,a large number of production and operation enterprises in China involve a wide range of types,systems,and distribution.Moreover,due to the generally low level of safety management of enterprises,a large number of hidden dangers may lead to frequent production safety accidents.At present,the main way of accident control and prevention is post-emergency treatment.Based on the characteristics of a machine-processing enterprise and the actual production situation,this paper analyses the data of safety production inspection of a machine-processing enterprise.According to the accident level which may be caused by hidden dangers in the enterprise,the risk level of a possible accident in the enterprise corresponds to the risk level of hidden dangers in the inspection cycle,and is divided into four categories: high,medium,low and extremely low.The fault tree model is established by selecting explosion and electric shock accidents in mechanical processing enterprises.Then the image mapping from fault tree model to Bayesian network is processed.The hidden danger risk level of root node is analyzed and predicted by Markov chain.Then the risk level of leaf node(top event)can be obtained by Bayesian network model,and the root can be obtained.The posterior probability distribution of node hidden danger events is conducive to putting forward pertinent preventive and control measures.It is of great significance for enterprises to take proactive preventive measures instead of passive safety management forms for their own safety management and accident prevention.Finally,using the analysis method of this paper,the safety production situation of the enterprise in the next five months is analyzed and predicted,and the risk level of two kinds of accidents that may occur in the enterprise is obtained.According to the analysis results and the posterior distribution of the risk level probability of the bottom event,reasonable countermeasures are put forward. |
PDF Full Text Request