Reliability Assessment Of Complex Systems Considering Common Cause Failure

With the development of science and technology,accurate reliability assessment is needed to control the reliability level reasonably for gradually complex system. The phenomenon of common cause failure has become an important factor which cannot be ignored. At the same time, there often exists zero-failure in the reliability test. In this context,from the viewpoint of system and degradation theory, this paper has studied the method of complex system reliability assessment considering common cause failure by combining with Bayesian network.In view of the test with zero-failure data, we could use the degradation theory and predict the life of the unit based on the main performance trends. This method can greatly shorten the cycle of reliability work, improve the assessment precision and efficiency. In case study, we analyzed the effects on the assessment results by using different life distributions: Exponential distribution, Normal distribution and Weibull distribution. Using the theory of systems engineering to solve the problem of small sample or no sample on system level by converting the information of the bottom unit into the system information.The system evaluation method of the systems composed by pass-fail components and the systems whose components are the units which in same life distribution type were summarized. For hybrid system composed of different distribution types, were studied the data conversion method and the Carlo Monte method. The MC method based on Fiducial point and the bootstrap method were using in the case, which proved the validity of this method. In order to add the common factor into the model, Bayesian network was introduced to model common cause failure system. Three kind of Bayesian network models were discussed in this paper,and using Monte Carlo method to solve the Bayesian network.The case shows that the impact of the common cause failure on the system reliability assessment cannot be ignored.In summary, this paper studied a framework for reliability assessment of complex system considering common cause failure. The framework for the conduct in reliability assessment of complex systems has a certain reference value.