Research On Reliability Of Pipelayer Hydraulic System Based On Universal Generation Function And Bayesian Network

The internal structure of large-scale electromechanical equipment systems and the relationship between units are becoming more and more complex,and it is particularly important to study the reliability of complex systems.Once the system or unit fails,it may cause serious social impact.The current complex system reliability model is not comprehensive enough to consider factors.The multi-state nature of the complex system in actual engineering,the state probability of each unit and the uncertainty of the state performance level information,the common cause failure between the units,and the working process should be considered.The establishment and analysis of the reliability model is more in line with the actual situation of the project.This dissertation aims at the characteristics of multi-state,uncertainty of sample information of state probability and state performance level of each unit,common cause failure,dynamic time series,and difference in the distribution of parameter random variables in complex systems,and establishes a complex multi-state system reliability model from multiple factors,and the system analysis is complex.Reliability issues of multi-state systems,improve the reliability theory system of complex multi-state systems,and provide theoretical support for actual engineering reliability.This dissertation introduces Dempster-Shafer(D-S)evidence theory,cloud model theory,common cause failure theory and Phase-type(PH)distribution theory on the basis of universal generation function and Bayesian network.Establish and analyze the reliability model of complex multi-state system from two aspects of mathematical structure and network structure.In view of the reliability of complex multi-state systems in consideration of different situations,corresponding reliability analysis methods are proposed.Evidence theory and cloud model theory solve the problem of uncertainty in the state probability information and state performance level of each unit in a complex multi-state system.Because PH has strong versatility,it is used to solve the problem of the difference in the residence time distribution of different unit states in complex multi-state systems.In solving the problem of the dynamic timing of complex polymorphic systems,a discrete-time Bayesian network method of evidence is adopted,and the reliability index of each time state is solved according to the different working time states of the system.For the common cause failure problem of complex multi-state systems,the single common cause failure group and the mutual exclusion and overlap between the multiple common cause failure groups are considered,and the weight influence vector method,? factor model method,and ? factor model method are adopted.And improve the ? factor model method for analysis.The comprehensive application of the above methods makes the reliability theory system of complex multi-state systems more perfect and more in line with actual engineering conditions.The hydraulic system of the pipelayer is a large-scale electromechanical equipment system,which conforms to the characteristics of a complex multi-state system.This dissertation analyzes the reliability of each hydraulic system of the pipelayer according to the method proposed in this dissertation and verifies the feasibility of the method.By comparing the reliability indexes of each hydraulic system of the pipelayer with independent failure and common cause failure,the influence of common cause failure on the reliability of each hydraulic system of the pipelayer is analyzed.According to the actual situation in the work of the pipelayer,corresponding maintenance measures are formulated for the important units in each hydraulic system to improve the reliability of the system.The theories and methods proposed in this paper can provide theoretical and technical support for actual projects.The dissertation has 47 figures,39 tables,and 245 references.