Nuclear Techniques ›› 2019, Vol. 42 ›› Issue (6): 60601-060601.doi: 10.11889/j.0253-3219.2019.hjs.42.060601

• NUCLEAR ENERGY SCIENCE AND ENGINEERING • Previous Articles     Next Articles

Reactor loading pattern optimization using improved differential evolution

Hui DING1,2,Guangyao SUN1,Bin WU1,Lijuan HAO1(),Yican WU1   

  1. 1. Key Laboratory of Neutronics and Radiation Safety, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences, Hefei 230031, China
    2. University of Science and Technology of China, Hefei 230027, China
  • Received:2019-01-14 Revised:2019-04-08 Online:2019-06-10 Published:2019-06-18
  • Contact: Lijuan HAO
  • About author:DING Hui, male, born in 1993, graduated from North China Electric Power University in 2014, doctoral student, focusing on intelligent nuclear design
  • Supported by:
    Supported by National Key R&D Program of China (No.2018YFB1900601), Strategic Priority Research Program of Chinese Academy of Sciences (No.XDA22010504), Key Research Program of Chinese Academy of Sciences (No.ZDRW-KT-2019-1-0202), Project of HIPS (No.KP-2017-19), Project of Chinese Academy of Sciences (No.XXH13506-104), Young Elite Scientists Sponsorship Program by CAST(No.2017QNRC001)

Abstract: Background

Core loading pattern optimization is an important task in core design. It features huge search space, a large number of design variables and strong nonlinearity. Traditional optimization methods are difficult to find the optimal solution.


This study aims to optimize reactor loading pattern by using improved differential evolution method.


Based on a new evolutionary algorithm called “differential evolution algorithm”, the TDDE (DE with TSP and dynamic strategy) method for loading optimization was developed to minimize the power peaking factor (PPF) under the constraint that k eff was not reduced. The improved mutation and crossover operator was used to optimize the discrete variables. In addition, SuperMC was used to calculate the reactor criticality.


The results of above method applied to the optimization core refueling pattern for WestingHouse PWR Cycle 1 showed that the optimized refueling pattern of TDDE reduces the pf by 21.3% without the decrease of k eff.


The TDDE method is thus a very promising method for optimizing core loading pattern.Key words Differential evolution, TDDE, Reactor loading pattern

CLC Number: 

  • TL329+.2