Font Size: a A A

Reliability Analysis Based On Bayesian Network Of EMU Transmission System

Posted on:2023-02-22Degree:MasterType:Thesis
Country:ChinaCandidate:Y QiuFull Text:PDF
GTID:2542307145966649Subject:Vehicle engineering
Abstract/Summary:PDF Full Text Request
Transmission system is an important part of EMU,and the reliability index of EMU directly affects the stability and safety of EMU.Therefore,it is of great practical engineering significance to analyze the reliability of the transmission system and improve the reliability and maintenance efficiency of the transmission system.In this paper,an improvement is proposed on the basis of Bayesian network reasoning method.The Bayesian network is divided into single-connected network and multi-connected network,and the reasoning complexity is reduced by decomposing the Bayesian network first and then reasoning.Taking the transmission system of EMU as the research object,the Bayesian network of transmission system fault mode is established,and the reliability of transmission system is analyzed by using the improved Bayesian network reasoning method.The details are as follows:Firstly,an improved method of Bayesian network reasoning process is proposed and verified.Bayesian networks are divided into single-connected networks and multi-connected networks.The single-connected network uses the tables of probability distributions to reasoning.The multi-connected networks are decomposed and reasoned layer by layer from bottom to top.Then the inference of the whole Bayesian network is carried out by combining the inference results of single-connected networks and multi-connected networks.By comparing the reasoning process and results with the traditional bucket elimination method,the accuracy of the improved reasoning method is verified,the complexity of the reasoning process is reduced and the reasoning efficiency is improved.Secondly,the transmission system is analyzed based on improved Bayesian network reasoning method.By analyzing and summarizing the structural characteristics and fault modes of EMU transmission system,the Bayesian network of transmission system fault modes is established.Through the fuzzy comprehensive evaluation of the occurrence probability of fault mode,the prior probability of each node of Bayesian network of transmission system is obtained.The improved inference method of Bayesian network is used to infer Bayesian network of transmission system from two directions,and the probability of system failure is obtained,as well as the probability of the system failure caused by a certain failure mode in the case of system failure is obtained.Finally,the reliability of the transmission system is analyzed based on the inference results of Bayesian network.Through the posterior probability of the transmission system,the components prone to failure of the system are found,and the critical importance is used to analyze the impact of the components of the transmission system on the reliability of the system.Then the posterior probability and critical importance of the system are compared to find out the important components and weak parts of the system,and provide suggestions for the reliability design and maintenance of the system.The reliability allocation of the system is carried out according to the Bayesian network of the transmission system.The reliability index of each component is determined,which provides reference for the improvement design of the system.
Keywords/Search Tags:Transmission system, Bayesian network, Reliability analysis, Fuzzy theory, Importance degree
PDF Full Text Request
Related items