RELAP5 / MOD3.2 analysis of INSC standard problem INSCSP - V4 : investigation of heat transfer for partly uncovered VVER-1000 core at the test facility KS (RRC K1).

Tentner, A. ; Ahrens, J. W.

Argonne National Laboratory

关键词: Heat Transfer;    Radiation Protection;    99 General And Miscellaneous//Mathematics, Computing, And Information Science;    Computerized Simulation;    Sensitivity;   

DOI  :  10.2172/822580
RP-ID  :  ANL/RAE/RP-108103
RP-ID  :  W-31-109-ENG-38
RP-ID  :  822580

美国|英语

来源: UNT Digital Library

【 摘 要 】

The RELAP5/MOD3.2 computer program has been used to analyze a series of tests investigating heat-transfer from a partly uncovered VVER-1000 core in the KS test facility at the Russian Research Center ''Kurchatov Institute'' (RRC-KI). The analysis documented represents VVER Standard Problem 4 in Joint Project 6, which is the investigation of Computer Code Validation for Transient Analysis of RBMK and VVER Reactors, between the United States and Russian International Nuclear Safety Centers. The experiment facility and data, RELAP5 nodalization, and results are shown for one of the six tests defined in Standard Problem 4. Only part of the data was analyzed due to our conclusion that the available experimental data is not sufficient to allow the modeling of the actual experiment sequence. The experiment initial conditions were reached through a series of transient processes, about which no quantitative information was available. This has required the modeling of an arbitrary computational transient, with the goal of reaching initial conditions similar to those observed during the experiment. The use of an arbitrary transient introduces many degrees of freedom in the analysis, i.e. initial computational values that influence the entire sequence of events, including the loop behavior during the experiment time window. Reasonable agreement between RELAP5 and the experiment data can be obtained by manipulating a number of initial computational values, including the liquid level in the fuel assembly model, the liquid level in the annular region, the quality of the saturated vapor in the voided loop regions, etc. Our study has focused on exploring the sensitivity of results to changes in these initial values which are not based on experimental information, but are selected with the goal of matching the experimentally observed behavior during the experiment time window. We have shown that changes in several initial arbitrary values can lead to similar changes in the calculated loop behavior. Other parameters exist which can directly influence the calculated results, which have been only minimally explored, or have not been explored at all due to lack of time. Because we are not modeling the entire sequence of events which led to the initial experiment conditions, it is not possible to assume that by matching a limited set of measured parameters at the beginning of the experiment time window we have reproduced all the initial actual conditions. We conclude that the analysis of these experiments, while providing a useful demonstration of the RELAP5 capabilities to describe the experimental sequence of events, cannot be used to draw quantitative conclusions about the adequacy of specific RELAP5 models.