Research On Dynamic Reliability Analysis Method Of Nuclear Power Plant System Based On CS-DFT

Nuclear safety is the premise and basis for the development of nuclear power.A large number of safety systems with redundant backups are used in nuclear power plants with obvious dynamic characteristics.Traditional fault tree analysis method has deficiencies in analyzing system dynamics.The dynamic fault tree analysis method based on the cut-sequence method has advantages in characterizing the dynamic characteristics and qualitative and quantitative analysis of the system.In this thesis,the dynamic fault tree method based on the cut-sequence method is applied to the dynamic reliability analysis of nuclear power plant safety system,which lays a foundation for better understanding of the dynamic characteristics of nuclear power plant safety system.First,based on the summary of the research status of dynamic fault tree analysis methods in China and abroad,the problems of dynamic fault tree modeling and qualitative and quantitative analysis of complex systems are proposed.The pre-processing and modular methods are used to establish a dynamic fault tree model of complex systems.The case analysis is compared with the Bayesian network analysis method.The method is described to illustrate the dynamic characteristics of the system and to obtain the advantages of the accident sequence.Secondly,the qualitative and quantitative analysis method for complex systems based on the cut-sequence method is studied.The analysis process and steps are summarized.The comparison with the Markov-chain method shows that the cut-sequence method can not only obtain the failure sequence of the accident and reveals the weak link of the system,but also avoid the problem of state space explosion and reduce the computational complexity.Finally,the nuclear power plant component cooling system is taken as the research object,and a modular system dynamic fault tree model is established.The quantitative sequence analysis is carried out by using the cut-sequence method,and the minimal cut sequence set of the component cooling system under non-LOCA is obtained.28 sequences of accidents leading to the top event were obtained to reveal the most dangerous link leading to the top event,and the top event's unreliability obtained after eliminating the impossible sequence of accidents is 8% lower than that obtained by the traditional fault tree analysis method.The analysis results show that the dynamic fault tree analysis method based on the cut-sequence method can avoid the shortcomings of the Markov-chain method and the traditional fault tree analysis method for complex systems with dynamic characteristics of nuclear power plants,and can describe the dynamic characteristics of complex systems,eliminate the sequence of accidents that are impossible,avoid the problem of state space explosion,reduce the computational complexity and get more accurate results.The research of the thesis lays a foundation for better understanding of the dynamic characteristics of nuclear power plant safety systems.