DEVELOPMENT OF INTERNAL FLOODING PSA FOR NEW BUILD UK GENERIC DESIGN ASSESSMENT

Transcription

1 DEVELOPMENT OF INTERNAL FLOODING PSA FOR NEW BUILD UK GENERIC DESIGN ASSESSMENT PSA 2017, Pittsburgh Richard Derrett-Smith Jacobsen Analytics Ltd.

2 OVERVIEW The UK Generic Design Assessment (GDA) Process UK ABWR Design Assessment Approach Assessment Insights Conclusions

3 UK GDA PROCESS Developed in 2007 following UK Government Energy Review in 2006 Objective is to facilitate regulator (ONR) input early in the permissioning process to improve regulator efficiency and reduce commercial risk for developers. Multi-step process in increasing detail: Step 1 - Preparation of design, safety case and security submissions Step 2 Fundamental (high level) design, safety case, security claims Step 3 Overall design, safety case and security claims review (Methods) Step 4 Detailed assessment of design, safety case and security (Assessment) (RP) Objective is to receive a Design Acceptance Certificate (DAC) to allow progression to the site-specific licencing phase.

4 UK ABWR DESIGN Gen III+ Boiling Water Reactor based on J- ABWR design J-ABWR operational at four sites in Japan Some design changes necessary for UK sites First planned construction at Wylfa in North Wales Site-licencing process for Wylfa about to commence All current new-build sites are on existing nuclear licenced sites Image produced from

5 ASSESSMENT PROCESS - 1 Internal Flooding PSA based on EPRI Guide (EPRI ) to Internal Flooding PRA Established 10-task process as outlined below: Tried and tested approach which meets ONR TAG expectations and ASME/ANS standard HLRs and SRs Some issues with application to plants in design however..

6 ASSESSMENT PROCESS - 2 Application of the method to plants in generic design assessment can face the following issues: Scope of the design under consideration not all plant areas considered General limitations in availability of detailed design information Synchronisation of deterministic (design-basis) assessments with probabilistic assessments General limitations in availability of operational procedures during normal operation and maintenance

7 ASSESSMENT APPROACH QUALITATIVE EVALUATION - PROBLEMS Principally an information gathering and initial screening activity to provide sufficient information to develop meaningful flood scenarios. As discussed, some gaps in available information: Spatial position of SSCs in flood areas difficult to measure spray impacts Flood height of SSCs difficult to measure submergence impacts Lack of flood source information for new buildings Drainage systems not finalised sump detection not possible, even spray scenarios could propagate Physical walkdowns not possible

8 ASSESSMENT APPROACH QUALITATIVE EVALUATION - SOLUTIONS Possible to make initial assumptions with conservative bias, e.g. Assume all spray susceptible equipment failed in a flood area given a flood event Use flood source information from equivalent rooms in surrogate building to overcome gaps, e.g. pump rooms, diesel generator rooms Use electronic model as surrogate for physical walkdown.

9 ASSESSMENT APPROACH - QUANTITATIVE EVALUATION - 1 Use of assumptions with conservative bias led to very high initial results need to consider opportunities for refinement of initial scenarios such as: Conduct detailed flood height calculations for risk-significant flood areas Use 3D CAD model to refine risksignificant HELB scenarios

10 ASSESSMENT APPROACH QUANTITATIVE EVALUATION - 2 The importance of operator actions both from an initiating and mitigating perspective was significant. Recognising credited operator actions can be adversely impacted by flooding events affecting credited instrumentation Crediting operator actions to isolate or terminate flooding events

11 ASSESSMENT CONCLUSIONS & INSIGHTS If plant-specific data is unavailable or limited, avoid use of overly conservative assumptions. Focus on uncertainty and sensitivity analysis to understand impact of assumptions on risk results. Distinguish between impact of conservatism and genuine design / procedural shortfalls Given it is a generic design assessment identify gaps in information that need to be addressed in a future update Compare key assumptions with equivalent deterministic assessment to ensure consistency of approach and insights

12 QUESTIONS AND COMMENTS? Richard Derrett-Smith Telephone: