Optimization of boiling water reactor loading pattern using two-stage genetic algorithm

Yoko Kobayashi, Eitaro Aiyoshi

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

A new two-stage optimization method based on genetic algorithms (GAs) using an if-then heuristic rule was developed to generate optimized boiling water reactor (BWR) loading patterns (LPs). In the first stage, the LP is optimized using an improved GA operator. In the second stage, an exposure-dependent control rod pattern (CRP) is sought using GA with an if-then heuristic rule. The procedure of the improved GA is based on deterministic operators that consist of crossover, mutation, and selection. The handling of the encoding technique and constraint conditions by that GA reflects the peculiar characteristics of the BWR. In addition, strategies such as elitism and self-reproduction are effectively used in order to improve the search speed. The LP evaluations were performed with a three-dimensional diffusion code that coupled neutronic and thermal-hydraulic models. Strong axial heterogeneities and constraints dependent on three dimensions have always necessitated the use of three-dimensional core simulators for BWRs, so that optimization of computational efficiency is required. The proposed algorithm is demonstrated by successfully generating LPs for an actual BWR plant in two phases. One phase is only LP optimization applying the Haling technique. The other phase is an LP optimization that considers the CRP during reactor operation. In test calculations, candidates that shuffled fresh and burned fuel assemblies within a reasonable computation time were obtained.

Original languageEnglish
Pages (from-to)119-139
Number of pages21
JournalNuclear Science and Engineering
Volume142
Issue number2
Publication statusPublished - 2002 Oct

Fingerprint

Boiling water reactors
Genetic algorithms
Control rods
Mathematical operators
Reactor operation
Hydraulic models
Computational efficiency
Simulators

ASJC Scopus subject areas

  • Nuclear Energy and Engineering

Cite this

Optimization of boiling water reactor loading pattern using two-stage genetic algorithm. / Kobayashi, Yoko; Aiyoshi, Eitaro.

In: Nuclear Science and Engineering, Vol. 142, No. 2, 10.2002, p. 119-139.

Research output: Contribution to journalArticle

Kobayashi, Yoko ; Aiyoshi, Eitaro. / Optimization of boiling water reactor loading pattern using two-stage genetic algorithm. In: Nuclear Science and Engineering. 2002 ; Vol. 142, No. 2. pp. 119-139.
@article{990c359d26224df1ba3c6caa21fc70c7,
title = "Optimization of boiling water reactor loading pattern using two-stage genetic algorithm",
abstract = "A new two-stage optimization method based on genetic algorithms (GAs) using an if-then heuristic rule was developed to generate optimized boiling water reactor (BWR) loading patterns (LPs). In the first stage, the LP is optimized using an improved GA operator. In the second stage, an exposure-dependent control rod pattern (CRP) is sought using GA with an if-then heuristic rule. The procedure of the improved GA is based on deterministic operators that consist of crossover, mutation, and selection. The handling of the encoding technique and constraint conditions by that GA reflects the peculiar characteristics of the BWR. In addition, strategies such as elitism and self-reproduction are effectively used in order to improve the search speed. The LP evaluations were performed with a three-dimensional diffusion code that coupled neutronic and thermal-hydraulic models. Strong axial heterogeneities and constraints dependent on three dimensions have always necessitated the use of three-dimensional core simulators for BWRs, so that optimization of computational efficiency is required. The proposed algorithm is demonstrated by successfully generating LPs for an actual BWR plant in two phases. One phase is only LP optimization applying the Haling technique. The other phase is an LP optimization that considers the CRP during reactor operation. In test calculations, candidates that shuffled fresh and burned fuel assemblies within a reasonable computation time were obtained.",
author = "Yoko Kobayashi and Eitaro Aiyoshi",
year = "2002",
month = "10",
language = "English",
volume = "142",
pages = "119--139",
journal = "Nuclear Science and Engineering",
issn = "0029-5639",
publisher = "American Nuclear Society",
number = "2",

}

TY - JOUR

T1 - Optimization of boiling water reactor loading pattern using two-stage genetic algorithm

AU - Kobayashi, Yoko

AU - Aiyoshi, Eitaro

PY - 2002/10

Y1 - 2002/10

N2 - A new two-stage optimization method based on genetic algorithms (GAs) using an if-then heuristic rule was developed to generate optimized boiling water reactor (BWR) loading patterns (LPs). In the first stage, the LP is optimized using an improved GA operator. In the second stage, an exposure-dependent control rod pattern (CRP) is sought using GA with an if-then heuristic rule. The procedure of the improved GA is based on deterministic operators that consist of crossover, mutation, and selection. The handling of the encoding technique and constraint conditions by that GA reflects the peculiar characteristics of the BWR. In addition, strategies such as elitism and self-reproduction are effectively used in order to improve the search speed. The LP evaluations were performed with a three-dimensional diffusion code that coupled neutronic and thermal-hydraulic models. Strong axial heterogeneities and constraints dependent on three dimensions have always necessitated the use of three-dimensional core simulators for BWRs, so that optimization of computational efficiency is required. The proposed algorithm is demonstrated by successfully generating LPs for an actual BWR plant in two phases. One phase is only LP optimization applying the Haling technique. The other phase is an LP optimization that considers the CRP during reactor operation. In test calculations, candidates that shuffled fresh and burned fuel assemblies within a reasonable computation time were obtained.

AB - A new two-stage optimization method based on genetic algorithms (GAs) using an if-then heuristic rule was developed to generate optimized boiling water reactor (BWR) loading patterns (LPs). In the first stage, the LP is optimized using an improved GA operator. In the second stage, an exposure-dependent control rod pattern (CRP) is sought using GA with an if-then heuristic rule. The procedure of the improved GA is based on deterministic operators that consist of crossover, mutation, and selection. The handling of the encoding technique and constraint conditions by that GA reflects the peculiar characteristics of the BWR. In addition, strategies such as elitism and self-reproduction are effectively used in order to improve the search speed. The LP evaluations were performed with a three-dimensional diffusion code that coupled neutronic and thermal-hydraulic models. Strong axial heterogeneities and constraints dependent on three dimensions have always necessitated the use of three-dimensional core simulators for BWRs, so that optimization of computational efficiency is required. The proposed algorithm is demonstrated by successfully generating LPs for an actual BWR plant in two phases. One phase is only LP optimization applying the Haling technique. The other phase is an LP optimization that considers the CRP during reactor operation. In test calculations, candidates that shuffled fresh and burned fuel assemblies within a reasonable computation time were obtained.

UR - http://www.scopus.com/inward/record.url?scp=0036823855&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036823855&partnerID=8YFLogxK

M3 - Article

VL - 142

SP - 119

EP - 139

JO - Nuclear Science and Engineering

JF - Nuclear Science and Engineering

SN - 0029-5639

IS - 2

ER -