OPTIMIZATION OF FUEL RELOADING PATTERNS FOR A RESEARCH REACTOR BY SIMULATED ANNEALING

Abstract

This article presents results obtained from a research into an application of simulated annealing method to the in-core fuel reloading pattern optimization for a research reactor. The decision variable of the optimization problem is a fuel reloading pattern for the next cycle after the present cycle finishes. The objective function maximizes the effective multiplication factor keff at the beginning of cycle while it is established to include an important safety paramater – the power peaking factor, in search process. A procedure for searching the optimal solutions was formed and a computer code was developed in the Fortran language running on PCs. Nuclear safety parameters for the optimization problem are provided from the results of the multigroup neutron diffusion theory computation program CITATION. A sample calculation was performed to find the optimal fuel reloading patterns for the second cycle of the Dalat research reactor and the results are presented in this article.