JISE


  [ 1 ] [ 2 ] [ 3 ] [ 4 ] [ 5 ] [ 6 ] [ 7 ] [ 8 ] [ 9 ] [ 10 ] [ 11 ] [ 12 ] [ 13 ] [ 14 ] [ 15 ]


Journal of Information Science and Engineering, Vol. 33 No. 1, pp. 37-50


An Enhanced Multi-objective Group Search Optimizer Based on Multi-producer and Crossover Operator


XIANG-WEI ZHENG1,2, XIAO-MEI YU1,2, YAN LI1,2 AND HONG LIU1,2
1
School of Information Science and Engineering
Shandong Normal University
Ji’nan
, 250014 P.R. China
2Shandong Provincial Key Laboratory for Distributed Computer Software Novel Technology
Ji’nan
, 250014 P.R. China
E-mail: xwzhengcn@163.com


    In order to enhance the convergence ability of multi-objective group search optimizer and improve solution distribution of non-dominated Pareto set, we put forward a novel multi-objective group search optimizer based on multiple producers and crossover operator of genetic algorithm (MCGSO) in this paper. The producer of MCGSO is extended from one to multiple ones, which explores more solutions and improves solution distribution of non-dominated Pareto set. For the purpose of preventing a local optimal solution, the metropolis rule of simulation annealing algorithm is introduced into the search pattern of producers. Rangers’ search strategies and crossover operator are combined to enhance algorithm’s ability to find new solutions and expand the range of non-dominated optimal set. Experimental results on DZTL serial benchmark functions demonstrate that MCGSO can effectively and efficiently solve multi-objective optimization problems compared with other similar multi-objective evolutionary algorithms.


Keywords: multi-objective optimization, group search optimizer, multi-producer, crossover operator, Pareto set

  Retrieve PDF document (JISE_201701_03.pdf)