Verification of the EUCLID/V2 Code Based on Experiments Involving Destruction of a Liquid Metal Cooled Reactor's Core Components

摘要

Abstract— The article presents the results obtained from verification of the EUCLID/V2 coupled code developed at the Nuclear Safety Institute of the Russian Academy of Sciences, which is intended for analysis of accident conditions in liquid metal cooled fast reactors. The additional capabilities available in the code in comparison with its first version include, in particular, analysis of individual equipment components in the 3D approximation, consideration of the transport of fission products and corrosion products in the coolant and in the nuclear power plant buildings, and also analysis of severe accidents in a fast reactor. The article presents the code verification results and assessment of its applicability for analysis of accidents involving destruction of fuel pins and the reactor core. The verification was carried out against the data obtained at experimental facilities and from analytic tests. Information about the key experiments used to validate the code is briefly outlined. In particular, data of experiments carried at the Oak Ridge, Argonne, and Sandia National Laboratory, the United States; at the National Nuclear Center of the Kazakhstan Republic; and on the test bench at the Nizhny Novgorod State Technical University (NSTU) in Russia are used. Modules of the coupled code EUCLID/V2 integrated code verification matrix are given. The errors of calculating the parameters most important for analysis of an accident’s consequences evaluated using the EUCLID/V2 code are proven with due regard to the computation and experimental results. The ranges of parameters within which the code has been verified are determined. The uncertainty and sensitivity of computation results are also analyzed based on the results from simulating certain experiments. Factors having the main influence on the computation results are determined. It is shown that the computation results are consistent with the experimental results subject to the input data uncertainty.