System Reliability Evaluation Based On Copula Function And Decision Diagram

With the progress of society and the rapid development of science and technology,equipment systems in various fields such as communications,weaponry,electric power,computer networks,etc.are showing increasing performance requirements,and the development trend of more complex and dependent structure,which has an impact on the reliability of these equipment systems.The traditional reliability analysis methods often only consider independence component failures in the system,ignoring the analysis of correlated failures among the system components.Failure to consider the correlated failures will lead to discrepancies between the reliability evaluation results and the actual situation,resulting in certain hidden dangers(e.g.,misleading the reliability design optimization and maintenance activities).Thus,it is significant to model and analyze the correlated failures to improve the accuracy of the system reliability assessment results.This thesis mainly uses the decision diagram method to model and analyze the system reliability,uses the Copula function as a tool to describe the correlation,and uses single-phased mission systems and phased-mission systems as examples to investigate the effects of correlated failures on the system reliability.The main research content is divided into three parts:(1)Explore the influence of Copula on the reliability of different structural systems:For systems with different structures,the impacts of correlated failures on the system reliability can be different.The three common structures of the system are series systems,parallel systems and continuous k /n: F systems.We use the Copula function to derive the calculation formula of the system reliability for these three structures.We further explore the influence of the Copula function under different parameters on the reliability and average lifetime of the three systems with different numbers of components through calculation examples.The research in this part will provide theoretical support for research in subsequent chapters.(2)Reliability evaluation of single-phased mission system based on the Copula function:Both Bayesian networks and binary decision diagrams are used to model the reliability of the single-phased mission system.We verify that the binary decision diagram is a computationally more efficient method than the Bayesian network.Thus we combine the binary decision diagram and the Copula function for the reliability analysis of the single-phased mission system considering correlated failure.The influence of correlated failures on the reliability of the single-stage task system are also investigated.(3)Phased-mission system reliability evaluation based on the Copula function:Based on the research in the previous part,the decision diagram is used to model the reliability of the phased-mission system.It is verified through the model that the multi-valued decision diagram is easier to construct than the binary decision diagram,and the computation efficiency is higher when modeling a phased-mission system.Hence,we combine the multi-valued decision diagram and the Copula Function to model and analyze the reliability of a phased-mission system,and explore the influence of correlated failures on the reliability of the phased-mission system.We also extend the method to analyze the reliability of phased-mission systems subject to imperfect coverage and common cause failures.