Failure Risk Analysis Of Pulverized Coal Pressurized Conveying System Based On DFT-DBN

The pulverized coal pressurized conveying system is one of the important components of the dry pulverized coal pressurized gasification technology.The stability of its operating state directly affects the safety and production efficiency of the entire coal gasification unit.Due to the high complexity of the system and the correlation between the failure modes of devices and their dynamic characteristics,the traditional static risk assessment methods have been unable to meet their safety requirements.In order to truly reflect the dynamic logic relationship in the system,this thesis proposes a method of combining dynamic fault tree and dynamic Bayesian network to research the failure risk of the coal powder pressurized conveying system.Firstly,the technological process of the pulverized coal pressure conveying systemis was analyzed,the risk factors of four key devices such as bag filter,atmospheric powder silo,lock hopper system and feeding tank are identified,and a dynamic fault tree model is established on this basis.Then,based on the depth-first left-most traversal algorithm,the fault tree is divided into dynamic sub-tree and static sub-tree,and the failure mode that can cause cascading failure in the system is found using the cut-sequence method.Then,for the problem of missing fault data in actual engineering,fuzzy mathematics theory and expert evaluation method are used to obtain the approximate failure probability of the basic events of the dynamic fault tree.Finally,according to the conversion method from dynamic fault tree to dynamic Bayesian network,the dynamic risk assessment model of the pulverized coal pressure conveying system is established.With the powerful two-way reasoning ability of the dynamic Bayesian network,the failure risk analysis and fault diagnosis of the system when considering the time series,common cause failure,and maintenance factors are completed,and the maintenance interval and risk prevention and control measures of the system are given accordingly.The dynamic fault tree analysis method based on dynamic Bayesian network can effectively describe the dynamic risk evolution process of the pulverized coal pressure conveying system in different states,avoiding the errors caused by the analysis results due to neglect of time series,maintenance factors,and common cause failures.The research results have important theoretical significance and application value for coal chemical companies to further explore the potential hidden dangers of accidents and improve the scientificity of risk assessment.