Differential evolution algorithms applied to nuclear reactor core design

The differential evolution algorithm (DE) and a recently introduced variant, differential evolution with random localizations (DERL), are applied for the first time to a nuclear engineering optimization problem. This problem was previously solved with genetic algorithms, particle swarm optimization and Metropolis algorithms, and consists in adjusting several reactor cell parameters, such as dimensions, enrichment and materials, in order to minimize the average peak-factor in a three-enrichment-zone reactor, considering restrictions on the average thermal flux, criticality and sub-moderation. The results obtained by DE and DERL are compared against the published ones and both algorithms perform well, thus demonstrating their potential for other applications.