A study on the optimization of Integral Fuel Burnable Absorbers using the genetic algorithm based CIGARO fuel management system

Abstract

The CIGARO fuel management code, a genetic algorithm based optimization code developed by DeChaine was used in conjunction with deterministic studies to optimize Integral Fuel Burnable Absorbers (IFBA) assembly designs for Pressurized Water Reactors (PWRs). Penn State's IFBA assembly designs were compared to typical Westinghouse IFBA assembly designs of 64, 80, and 160 IFBA pins per assembly. The goal of this research is to load the core with various IFBA pin enrichments and various IFBA assembly designs to determine whether the IFBA pin configurations developed by the authors or those typically used by the vendor produce the best core performance when a real depletion is simulated. To further test the Penn State IFBA designs, a benchmark core loading modeled after Beaver Valley Cycle 10 was optimized with the CIGARO genetic algorithm code. When incorporated in a real depletion using CASMO-3 and SIMULATE-3 the Penn State and vendor's 64 IFBA) assembly designs with 2.5 wt% 10B and 1.875 wt% 10B, respectively, indicated good performance in terms of maximum core Normalized Power (NP) during depletion. All of the Penn State 80 IFBA assembly designs developed showed good performance in terms of maximum core NP during depletion. The vendor's 80 IFBA assembly with 1.25 wt% boron was very comparable to the Penn State 80 IFBA assembly design with 1.25 wt% boron in terms of NP performance. The core loadings generated by the CIGARO code provides evidence that optimizing BP placement directly by the Haling method predictions is not the most advantageous solution; however, the Haling method does provide a very good starting point to determine the number of IFBA assemblies needed in a core loading. This research advances the CIGARO optimization tool by including IFBA loading evaluations directly without any Haling Principle approximations.