.png){kind=icon}
.png){kind=icon}
Return
A vector angles-based many-objective particle swarm optimization algorithm using archive
.png){kind=icon}
PRE
AI
DOI：10.1016/j.asoc.2021.107299.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Because of the easy implement and fast convergence speed in single objective optimization problems, the particle swarm optimization algorithm has been extended to many-objective optimization problems. When designing a many-objective particle swarm optimization algorithm, archive maintenance strategy and selection of leaders are two crucial issues. To addressing these two problems, this paper proposed a vector angles-based many-objective particle swarm optimization algorithm using archive. In this algorithm, a new method was proposed to make the external archive getting a balance between convergence and diversity, based on the values of vector angles of each solution. Also, we designed a novel strategy based on vector angle and decomposition to select gbest and pbest from the archive to promote evolution of the population. Besides, the shift-based density estimation was used as the standard to clone elite solutions to strengthen the quality of the external archive. The experiment results indicate that our algorithm has competitive performance comparing with the six state-ofthe-art many-objective optimization algorithms on three widely used benchmarks DTLZ, WFG, and MaF. (C) 2021 Elsevier B.V. All rights reserved.
Keywords:
Many-objective optimization
Particle swarm optimization
Vector angle
Archive maintenance
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.6
Papers:
1.4W
Citations:
4.8W