A Novel Ranking-based Optimal Guides Selection Strategy in MOPSO

Publication Type:
Journal Article
Citation:
Procedia Computer Science, 2016, 91 pp. 1001 - 1010
Issue Date:
2016-01-01
Full metadata record
© 2016 Published by Elsevier B.V. A challenging issue with multi-objective particle swarm optimization (MOPSO) is the mechanism to select the optimal guides. This paper presents a new strategy based on ranking dominance and integrates into MOPSO. By using the ranking information and incorporating the chebychev distance of particle in objective space, we implement the selection of gbest and pbest simply and elegantly. On the basis of ranking, we propose a new maintenance strategy for updating the external archive which can obtain a more diverse and uniform distribution. Furthermore, a qualitative and quantitative analysis in terms of convergence analysis over some benchmarks is presented, providing a basis for conclusions about the proposed method. showing that the proposed method performs better than the adopted algorithms.
Please use this identifier to cite or link to this item: