Reliability Analysis Of Emus Sliding Plug Door System Based On Fuzzy Petri Net

With the great investment and development of China's high-speed railways,China's highspeed railways has been in the leading position in the world.At the same time,with the increase of high-speed rail mileage,how to ensure the safe and reliable operation of the EMU has been highly valued.The sliding plug door is an important component for passengers to get on and off,it is often unable to open or close due to human factors,external environmental factors and its own faults during the operation of the EMU,which affects the normal operation of the train,therefore,improving its reliability is very important to ensure the safe operation of the EMU.The main contents of this thesis are as follows:Firstly,this thesis expounds the construction and working principle of the sliding plug door system of the EMU of CRH5,and it sorts out the corresponding failure modes of each component of the system.Then it introduces the basic concepts of the fault tree analysis method,and a fault tree model of the sliding plug door system is preliminarily constructed according to the fault analysis of the sliding door system.Secondly,due to the uncertainty and fuzziness of fault data,the fault probability of basic events is uncertain,this thesis introduces a triangular fuzzy number to modify it,so that the basic failure probability is within an interval range,which is more consistent with the failure rate characteristics.It introduces the fuzzy mathematical theory to solve the problem of calculating and solving fuzzy numbers when using fuzzy fault trees for reliability analysis,which makes the final result more credible.Finally,the fault tree model of the sliding plug door system is quantitatively analyzed,and the importance of each component of the system is calculated.Thirdly,the traditional fault tree analysis method cannot solve both the dynamic characteristics of fault occurrence and propagation in the sliding plug door system and the characteristics of the diversity of the correlation between faults,as a result,this thesis introduces the fuzzy petri net model,which can not only describe it but also reduce the difficulty in solving the system's top event occurrence probability.Firstly,it gives the transformation process of fault tree to fuzzy petri net model.The transfer of fuzzy petri net token is used to describe the occurrence and propagation of system faults,and the validity and accuracy of the token transfer process are verified by combining the inference method of association matrix,which not only lays the foundation for the reliability analysis,but also improves the fault diagnosis efficiency of the system.Then,the uncertainty of the correlation between faults is solved by using the definition of transition confidence in fuzzy petri nets.The reliability of the system is obtained by the forward inference algorithm of the fuzzy petri net,and the importance of each component of the sliding plug door system is obtained.Finally,the results obtained by the fuzzy fault tree analysis method and the fuzzy petri net analysis method are compared,which proves that the results obtained by the fuzzy petri net analysis method are more reliable.Finally,because of the parameters of the rule generated in the fuzzy petri net are based on the statistical analysis results of the basic data and the comprehensive consideration of expert knowledge,there is a certain degree of subjectivity and ambiguity,which leads to inaccurate analysis results.Therefore,the particle swarm optimization algorithm is introduced to optimize the parameters of the fuzzy petri net model,which not only improves the generalization ability of the fuzzy Petri net,reduces the dependence of the model parameters on expert experience,but also improves the accuracy of the model parameters and makes the results obtained by the model more accurate.