A Novel Diversity Guided Particle Swarm Multi-objective Optimization Algorithm

Zhiyong Li, Ransikarn Ngambusabongsopa, Esraa Mohammed, Ndatinya Eustache -International Journal of Digital Content Technology and its Applications 2011 · DOI

Abstract

This paper presents a mu lti-objective d iversity gu ided Particle Swa rm Op timization a pproach named MOPSOAR wh ich increases di versity perfo rmance of multi-objective Parti cle S warm optimization by u sing A ttraction a nd Re pulsion (AR) m echanism. AR m echanism us es a d iversity measure to control the swar m. Being attractive and repulsive wil l help to overcome t he problem of premature c onvergence. AR mechanism together wi th cr owding di stance co mputation an d mutation operator maintains the diversity of non -dominated set in external archive. The approach is verified by several te st f unction exper iments. Results demo nstrate t hat the p roposed approach i s highly competitive in distribution of non-dominated solutions but still keeps convergence towards the Pareto front.
BibTeX
@article{li2011novel,
  author = {Zhiyong Li and Ransikarn Ngambusabongsopa and Esraa Mohammed and Ndatinya Eustache -},
  title = {{A Novel Diversity Guided Particle Swarm Multi-objective Optimization Algorithm}},
  journal = {International Journal of Digital Content Technology and its Applications},
  year = {2011},
  doi = {10.4156/jdcta.vol5.issue1.29},
}

Loading related content...

Figure