System reliability assessment method and device based on DFT qualitative and quantitative analysis

Abstract

An example embodiment in the invention provides a safety system reliability assessment method and device for nuclear power plants and the like, so that system-level DFT can be qualitatively and quantitatively analyzed quickly, accurate qualitative and quantitative analysis results can be given, and reparable or unreparable system can be analyzed. In the qualitative analysis, the system is modeled to be a dynamic fault tree DFT comprising dynamic module branches; the DFT dynamic module branches are converted into FT comprising dynamic events; the FT is converted into a binary decision diagram BDD comprising dynamic nodes; and the dynamic node BDD is utilized to determine accident sequence of top events of the DFT dynamic module branches, with the accident sequence being as qualitative information for system reliability assessment. In the quantitative analysis, a numerical integration method based on dynamic module level MCS is provided; and on the basis of independent random variables and by means of part sequence failure diagrams, an integration expression of a common module level MCS is educed and quantitative calculation is carried out.

Description

technical field [0001] Exemplary embodiments of the present invention generally relate to reliability assessment of safety systems, and specifically relate to a system reliability assessment method and device based on qualitative and quantitative analysis of Dynamic Fault Tree (DFT, Dynamic Fault Tree). Background technique [0002] In order to ensure the design balance and operation safety of industries that need to ensure safety, such as nuclear power plants, the reliability assessment of safe operation systems, especially important safety systems, has always been the focus and difficulty of research by technicians in the technical field. [0003] Taking nuclear power plants as an example, the traditional probabilistic safety analysis (PSA, Probabilistic Safety Analysis) of nuclear power plants is based on the static fault tree (SFT, Static Fault Tree) modeling method, and by converting the fault tree into a logically equivalent binary Meta-decision diagram (BDD, Binary De...

AI Technical Summary

Problems solved by technology

[0006] ①The Markov model method is widely used in the reliability assessment of repairable or non-repairable systems in nuclear power plants, but this method requires the life and repair time of components to obey the exponential distribution, that is, to meet the Markov characteristic (time has no memory), and with the As the scale of the system increases, the Markov model method will produce a "state space explosion" problem, making the problem solving process complicated or even impossible;

[0007] ②The numerical integration method avoids the "state space explosion", and even allows the lifetime of components to obey arbitrary distribution in some special cases, but the current method is mainly for dynamic logic gates and cannot deal with repairable problems;

[0008] ③ Although Monte Carlo simulation overcomes the defects of the above two methods, it cannot be widely used for small probability events because the simulation cost is too high

[0009] However, for the qualitative analysis of DFT in nuclear power plants, there are still few literatures involving

Images