Research On Reliability Of Complex Systems Based On Causality Diagram

With the continuous development of society and the advancement of science and technology,the dependence of modern industry on reliability has become more and more obvious.Reliability analysis technology can be used to periodically inspect equipment,determine the inspection cycle in service,and quickly find out the fault source when the equipment fails,guarantee the normal operation of the system,so as to help staff to formulate emergency or preventive measures.In this paper,causality diagram is used to analyze and predict the reliability of the system,which combines the theory of fuzzy mathematics and BP neural network.The main work of this paper is as follows:(1)This paper briefly introduces the knowledge expression and calculation method of causality diagram,and proves the rationality and validity of applying causality diagram to system reliability analysis with an example.(2)Fuzzy numbers are introduced on the basis of causality diagram to describe uncertain information in the real world.Firstly,the trigonometric fuzzy number and its algorithm are briefly introduced.Then,the probability of events is described by using the fuzzy number,and the fuzzy gate operation and the fuzzy conditional probability are defined.Through the quantitative analysis of causality diagram,we can get the probability of fuzzy condition of each event,determine the event which has the greatest impact on the system failure and the main cause of the intermediate event,and then analyze the reliability of the system.Finally,in the process of implementation of this method,the operation steps of this method are illustrated by an example,and the results of other methods are compared to verify the feasibility of this method.(3)Combining causality diagram with BP neural network,a new reliability prediction method for complex systems is proposed,and the operation steps of the method are explained with concrete examples.Firstly,the reliability of the system is calculated by the method of causality diagram,and then the BP network model is established according to the characteristics of the system.According to the historical data of the reliability of the system changing with the use time,the established networkmodel is used to analyze and process the data and train the network,so as to predict the reliability of the system at a certain time in the future.