Development and Validation of the Approach to Prediction of Mass Fraction of Agglomerated Debris
Severe accident management strategy adopted in Swedish type Boling Water Reactors (BWR) relies on reactor cavity flooding for providing ex-vessel melt coolability and termination of the accident progression. It is assumed that the melt ejected into a deep water pool will fragment and form a coolable by natural circulation debris bed. Agglomeration of partially liquid debris and formation of non-porous “cake regions in the bed can negatively affect coolability of the bed. The goal of the present work is development and validation of approach to prediction of different modes of agglomeration and mass fraction of agglomerated debris in order to reduce uncertainty in the ex-vessel corium coolability analysis. The VAPEX code is used as computational vehicle for simulation of the fuel coolant interaction (FCI). A conservative-mechanistic approach to quantitative estimation of mass fraction of agglomerated debris is proposed. The data from the DEFOR-A (Debris Bed Formation and Agglomeration) experiment is used for validation of the approach. Results of sensitivity study demonstrate that conservative treatment of epistemic uncertainties in agglomeration phenomena creates sufficient margin and simulation data are enveloping the set of experimental data. Developed approach is also applied to prediction of mass fraction of agglomerated debris at prototypic plant-scale severe accident conditions. Main finding is that mass fraction of agglomerated debris reduces rapidly below the leading edge of the jet even if considerable conservatism in the analysis is employed.
Severe accident Debris formation Debris agglomeration Cake formation
Pavel Kudinov Mikhail Davydov
Division of Nuclear Power Safety, Royal Institute of Technology (KTH)Roslagstullsbacken 21, SE-106 9 Electrogorsk Research and Engineering Center on Nuclear Power Plants Safety (EREC)Saint Constantine
国际会议
上海
英文
2021-2035
2010-10-10（万方平台首次上网日期，不代表论文的发表时间）