Multi-objective Optimization Algorithm Composed of Estimation of Distribution and Differential Evolution

TAO Xinmin,XU Peng,LIU Furong,ZHANG Dongxue
DOI: https://doi.org/10.3969/j.issn.1673-4785.201208035
2013-01-01
Abstract:In order to improve the ability of convergence and accuracy of a multi-objective optimization algorithm,a multi-objective optimization algorithm composed of estimation of distribution and differential evolution has been proposed.Both estimation of distribution algorithm and differential evolution algorithm will be used to generate particles of population.The generation method of each particle has been decided by using a selective factor,and proportion of the use of two algorithms according to the frequency of iterations.Utilizing an estimation of distribution algorithm to quickly locate in the initial search,and then differential evolution algorithm was used for accurately conducting searches.The variation factor of differential evolution algorithm was improved,and a variable variation factor also was defined and used to control the range of variation of differential evolution algorithm in different search periods.Four test functions were used to evaluate the performance of the proposed algorithm,and next the proposed algorithm was compared with nondominated sorting genetic algorithm II(NSGA-II) and regularity model-based multiobjective estimation of distribution algorithm(RM-MEDA).The experimental results show that the proposed algorithm displayed a good convergence,diversity performance,and the stable effects.
What problem does this paper attempt to address?