Research On Failure Diagnosis Method Of Chemical Equipment Based On Fault Tree

Due to the special nature of chemical raw materials and harsh process conditions,chemical industry is more dangerous,with more serious consequences when accident occurs.Chemical equipment failure is one of the important causes of the accident.The timely forecast and accurate detection of chemical equipment failure,can not only reduce the accident occurrence rate and improve the utilization rate of the equipment,but also have great significance for the realization of production safety and economic benefits.This paper first analyzes the characteristics of chemical equipment and its failure,according to the fuzzy and dynamic characteristics of the failure,and proposes to establish fuzzy dynamic fault tree model to diagnose chemical equipment failure.Then,the framework of fuzzy dynamic fault tree is established,and the core of this method is qualitative and quantitative analysis.In view of the shortcomings of the current qualitative and quantitative methods in calculation accuracy,efficiency and versatility of the model,combined with the high requirements of chemical equipment for fault diagnosis reliability and timeliness,this paper focuses on the qualitative and quantitative analysis methods of fuzzy dynamic fault tree of chemical equipment,and puts forward a systemic and complete diagnosis method.In the aspect of qualitative analysis,a qualitative analysis method of fuzzy dynamic fault tree based on improved sequence binary decision diagram is presented.By introducing temporal logic,the dynamic logic gates are firstly converted into temporal logic expressions.Then,according to the deficiency of the traditional dynamic fault tree to sequence binary decision diagram conversion approaches,the paper puts forward the reverse conversion method based on the component connection method,and gives the corresponding conversion and simplification rules.Finally,An example is analyzed,which shows the proposed method can reduce the number of redundant nodes and redundant structures of sequence binary decision diagram,and improve the efficiency of fuzzy dynamic fault tree qualitative analysis.In the aspect of quantitative analysis,a high universality,accurate and efficient quantitative analysis method is put forward.The fuzzy dynamic fault tree is decomposed into fuzzy static subtree and fuzzy dynamic subtree.Then based on minimum cut set and minimal cut sequence,quantitative analysis models of fuzzy static subtrees and fuzzy dynamic subtree are established.In order to reduce inaccurate calculation results caused by diffusion of data fuzziness,the concept of weak t-norm is introduced into the two subtrees.Finally,the calculation formula of the minimum cut set,cut sequence and the bottom event importance are given,and the process of fault diagnosis of chemical equipment is analyzed.Finally,the proposed method is applied to the fault diagnosis of a centrifugal compressor system,and the key part,the weak link and the fault diagnosis process of the system are obtained,verifying the feasibility and effectiveness of this method.