Study And Application Of Risk-informed Safety Margin Characterization Methodology

Nuclear power plants in China and around the world are facing the problems of power plant aging,operation extension and design change.The current safety analysis methods,including deterministic methodology and probabilistic methodology,have their own limitations.In order to help nuclear power plant owners to make more economic and safety decisions,and to provide more pragmatic regulatory approvals to nuclear power regulatory authorities.Through an in-depth theoretical research on the risk-informed safety margin characterization(RISMC)methodology,this paper finally and successfully develops and conducts a RISMC-based approach-computational risk assessment(CRA).This method is an effective way to integrate the traditional probability safety assessment(PSA)and best estimate plus uncertainty(BEPU)analysis to conduct safety analysis and evaluate risk-informed safety margin,especially towards the small power uprate of NPP.This paper focuses on the risk response of the CRA method applied to the system with small power uprate.By using this method,the traditional PSA model is improved and upgraded.After analyzing the significant probability sequences of a traditional three-loop PWR NPP SBO accident,considering and quantifying the uncertainty of the epistemic parameters and the stochastic parameters,reasonable and effective reduction of sampling intervals and sample numbers improves the efficiency and sensitivity of the entire risk assessment.In the end,not only the core damage frequency under the new method is calculated,but also the impact of small power uprate on plant risk is evaluated.The main content of this paper includes:(1)This paper expounds the requirements of the methodology for the next generation safety analysis,the required elements and the expected goals.Guided by the risk-informed safety margin characterization(RISMC)methodology,the combination of the deterministic methodology BEPU and probabilistic methodology PSA makes the advanced computational risk assessment(CRA)methodology and basic implementation steps.The methods that need to be integrated include(1)PSA modeling techniques,(2)dynamic analogy and analysis techniques for nuclear power plant accidents,(3)uncertainty analysis techniques,and(4)efficient sampling techniques and data statistics techniques.(2)The core damage frequency of two probabilistic significant sequences(PSSs)in SBO accident of a typical three-loop PWR NPP was re-analyzed with CRA method.The whole assessment includes modeling of real power plants,refining the key heading of the traditional PSA model(refining the success criteria into three parts:surely success,surely failure and conditional success),sampling both power plant status parameters and stochastic parameters randomly and quantifying the impact of uncertainty on the outcome.In general,on the basis of fully implementing the reality and quantifying the uncertainty of the two type parameters,compared with the calculation results of the traditional PSA,the CRA method can effectively reduce the calculated value of the core damage frequency,thus excavating a larger safety margin.(3)The CRA method is used to quantify the impact of a small power uprate successfully on the core damage frequency of station blackout in a typical pressurized water reactor.In the CRA method,when a small power uprate(5%)occurs in the power plant,t_min and t_max will shift to the left and the probability of surely failure increases.In addition,the conditional exceedance probability within the sampling range also increase,eventually resulting in a higher probability of failure in the success criteria,and thus the risk of a small power uprate can be quantified.In the end,two PSSs will be combined and discussed.Before the power uprate,the core damage frequency is 1.234E-5/year;after the power uprate of 5%,the core damage frequency increased to 1.389E-5/year.It means that the risk has increased by 12.56%.It is worth to know that the risk impact of this limited power uprate on power plants cannot be quantified with the traditional PSA methods.This paper develops a new set of safety analysis methods guided by the RISMC methodology with the next generation of analytical capabilities,which are both leading research in the world and in China.Moreover,this paper makes the theoretical prospects on next generation analysis capability into reality.And this method not only provides a pragmatic decision-making basis for power plant owners,but also provides new ideas and methods for licensing audits for nuclear power regulatory departments.