SOPRAG: a system for boiling water reactors reload pattern optimization using genetic algorithms

Genetic Algorithms (GA) are used in combination with the steady state nodal core simulator PRESTO-B to create a system for the optimization of reload patterns for Boiling Water Reactors (BWR). The system uses the basic GA operators, crossover, mutation and selection over the loading pattern (LP) represented by a combination of fresh and burned fuel assemblies, as well as an objective function taking into account cycle length and radial peaking factor, to obtain improved loading patterns compared with real BWR loadings. The system takes advantage of the efficient quarter core two dimensional (2D) calculations, using the Haling technique to perform thousands of LPs evaluations and obtain the better candidates in a reasonably computer processor (CPU) time.