The Studies Of Methodology And Applications For Common Cause Failure Analysis Of Probabilistic Safety Assessment For Third Qinshan Nuclear Power Plant

This thesis is to study the Common Cause Failure (CCF) anlysis, that is one of the important technical subjects of Probabilistic Safety Assessment (PSA) for Nuclear Power Plant. The functions of several devices or components can fail due to single shared cause. Common Cause Failure is a subset of dependent failure, where simultaneous or near simultaneous multiple failures result from a single shared cause. In the design of safety related systems of NPPs, redundancy is general applied for the purpose to insure the systems can meet their safey and reliability requirements. It meas the failure of the system is due to the failure of all or multiple identendial components which execute the same function. In the early stage of reliability engineering to calculate the reliability of a system with redundancy design, it was generally assumed the failure of identical components is independent, but according to the long time accumulation of industry experience, it is discovered the contribution of CCF of indential components to the globel system failure is much larger than the contribution of independent failure to the global system failure. For the reason, CCF is considered as one of the important technical subjects of PSA, and much more attention has been put on it.This thesis has performed deeply research on the theory of CCF and dependent failure. The thesis takes Shutdown System #2 of Third Qinshan NPP (TQNPP) as one example to construct the fault tree model to analysis the reliability of the system, and performes the comparesion analysis and evalution of Unified Partial Method (UPM) and Alpha-Factor Model. Additionally, this thesis performes the comparesion analysis and evalution to the CCF parameters of two kinds of safety significant equipments, Standby Diesel Generators, Emergency Diesel Generators and Emergency Core Cooling system Pumps by Unified Partial Method (UPM) and PWR generic CCF parameters. It is the first time to perform this kind of comparesion analysis and evaluation in international CANDU PSA fields. It is indicatd the following conclusions through the comparesion analysis and evaluation. UPM has performed appropriate qualitative analsis and quantitative calculation for CCF, and the method of UPM has properly incooperated NPP engineering practice. The calculation results of UPM is similare to the results by using generic Alpha factor parameters, but it is lack of qualitative CCF analsis by using the Alpha factor model. The application of Alpha factor model is just one method of quantitative calculation for CCF, and it is lack of application value from engineering points and is difficult to incooperate the real situation of TQNPP to CCF analysis. For the reason that UPM is properly incooperated engineering practice to CCF analysis, it is suggested to apply UPM to perform CCF analysis for PSA of TQNPP. Based on UPM methodology and TQNPP PSA engineering practice, through the research of the thesis, it is developed the executive procedure of CCF analysis for TQNPP PSA application project. The executive procedure will systematically solve the previous problem of without CCF analysis in TQNPP PSA model.As one of the technical subjects of TQNPP PSA application project to performe CCF methodology research and to determine the work precept, this thesis has taken the important role for the smooth development of TQNPP PSA application project.