Research On Reliability Analysis And Model Of Ship Nuclear Power Plant Based On Bayesian Network

Compared with other renewable energy sources,nuclear energy is superior,cleaner,and can stably and continuously realize large-scale supply so far.However,due to the speciality of the physical and chemical properties of nuclear fuel,the consequences of the leakage accident are disastrous.Therefore,it is still necessary for us to pay attention to safety problem while we advance the application of nuclear energy.Compared with terrestrial nuclear power plants,the nuclear-powered ships work in more complex marine environment and are more likely to experience accidents.The above factors require higher reliability of nuclear-powered ships.The nuclear power plant's second-loop system is an important part of the entire power system,which can provide the mechanical energy required for navigation of the ship,the electric energy and fresh water used by the entire ship.The research on the reliability analysis of the second-circuit system of nuclear power plant has important theoretical significance and application value.The second loop system of civil marine nuclear power plant is complex.There are many components and many kinds of failure modes in each subsystem.Because of the special working environment of the second-loop system,the precise probability of each component is difficult to obtain.Using only traditional reliability analysis methods such as fault tree analysis,failure modes and effect analysis,the results are inaccurate.Therefore,this paper selects the reliability analysis model based on Bayesian network to complete the reliability modeling and analysis of the second-loop system of ship nuclear power plant.The main achievements of this master work are listed as follows:1.The research constructs Bayesian network based on FMEA,FTA and BNT.In order to realize the application of engineering,the structural matrix is used to represent the logical relationship of variables in FMEA,FTA and BNT.Aiming at the problem of the accuracy of prior probability,this article studies expert-opinion error judgments,quantification rules,and opinions synthesizing.Then a systematic expert prior reliability of the method is thus verified.2.Based on the analysis of the functions and components of the nuclear power plant's second-loop system,the reliability analysis method was used to complete the FMEA and FTA analysis.Then using the GARMS software to complete the system failure mode,the impact analysis and fault tree analysis,the above steps verify the correctness of the previous analysis results.3.A Bayesian network reliability analysis model of the second-loop system is constructed based on the results of fault tree,failure modes and effect analysis,and the simulation of the model is completed based on the BNT toolbox in Matlab.Through the analysis of the calculation results,this research identify the weak links of the second-loop system,identify the factors that have a greater impact on the system,and provide support for the maintenance decisions and maintenance plans of the second-loop system.4.Based on the analysis of the functions and requirements of the system,Visual Studio,Visual Basic and Matlab are used to design and realize the basic reliability information collection,prior probability processing and reliability analysis in the reliability analysis platform of the second loop.