Heuristic Rules in Genetic Algorithm for Fuel Management Optimization

The objective of this study was to develop a unique§methodology and a practical tool for designing§loading pattern (LP) and burnable poison (BP)§pattern for a given Pressurized Water Reactor (PWR)§core. Because of the large number of possible§combinations for the fuel assembly (FA) loading in§the core, the design of the core configuration is a§complex optimization problem. It requires finding an§optimal FA arrangement and BP placement in order to§achieve maximum cycle length while satisfying the§safety constraints. To solve this optimization§problem, a LP optimization package, GARCO (Genetic§Algorithm Reactor Code Optimization) code is§developed. This code is applicable for all types of§PWR cores having different geometries and structures§with an unlimited number of FA types in the§inventory. To reach this goal, an innovative GA is§developed by modifying the classical representation§of the genotype. To obtain the best result in a§shorter time, not only the representation is changed§but also the algorithm is changed to use in-core§fuel management heuristics rules. The improved GA§code was tested to demonstrate and verify the§advantages of the new enhancements.