Search

Abstract: The AVR is a pebble-bed type reactor that operated in Germany for 21 years and was closed down in December 1988. The AVR melt-wire experiments, where graphite spheres with melt wires of different melting temperatures were introduced into the core, indicate that measured pebble temperatures significantly exceeded temperatures calculated with the analysis codes available at the time. The reason for these discrepancies are often attributed to the special design features of the AVR, in particular the control rod “noses” protruding into the core, and to inherent features of the pebble bed reactor. In a previous study different possible bypass flows were investigated. This study investigates different modeling strategies and assumptions for the solution of the core neutronics. Due to the complexities specific to the AVR there is not currently a code system that can take into account the noses while simultaneously solving the burnup and diffusion equations in three dimensions. A number of modeling simplifications were therefore made in the historic analysis of the AVR. The aim of this study is to quantify the effects these different simplifications have on the results. This includes the effects of modeling the core neutronics, burn-up and thermo-hydraulics in both two and three dimensions, as well as other simplifications of the geometry. The comparison of the most realistic case to the one comparable to historic calculations show that the difference in temperature predicted with these two models can be as high as 300°C. The gas temperature distribution at the top of the core, where the maximum temperatures occur, is in fair agreement with the melt-wire experiment data even though few simplifications are introduced to the models and the power history has not been simulated. The results also serve as input to the final modeling strategy to repeat the modeling of the operational history of the AVR, which is planned for the future. [Copyright &y& Elsevier]