Detailed neutronic calculations of the AP1000 reactor core with the Serpent code

Abstract

In this work we present some validation results for reactor core modeling with the Serpent code performed for the first cycle of the AP1000 reactor. The comparison with reported values of the assembly k∞ for cold zero-power condition showed a discrepancy of 0.29%. The kef for full-core static and burnup calculations of the very heterogeneous AP1000 reactor core also presented good agreement with reported values. The kef for states with uniform fuel and moderator temperature distributions showed discrepancies below 0.91%. The boron worth curve obtained from burnup calculations with the Serpent code model results reproduced very well literature results despite using uniform temperature distributions in the modeling. In addition we discuss shadowing effects among burnable absorber rods (IFBA and Pyrex) and control rods which are, together with soluble boron, the control means throughout the first cycle. For instance, the presence of 9 Pyrex rods in an assembly decreased the average reactivity worth of one IFBA rod from 147 pcm to 33 pcm; and the presence of 28 IFBA rods in an assembly decreased the average reactivity worth of one Pyrex rod from 631 pcm to 277 pcm. The reactivity worth of a black control rod reduces about 20% when 28 IFBA rods are inserted in the fuel assembly.