Summary
International Symposium on Nonlinear Theory and Its Applications
2016
Session Number:A3L-B
Session:
Number:A3L-B-1
An Improved Multi-Objective Particle Swarm Optimization Using an Efficient $G_{best}$ Selection Method
Yuki Hasegawa, Takayuki Kimura, Kenya Jin'no,
pp.-
Publication Date:2016/11/27
Online ISSN:2188-5079
Summary:
The accuracy, the uniformity and the breadth of the Pareto-front are important index for the multi-objective optimization problems. However, the conventional multi-objective particle swarm optimization (MOPSO) considers only uniformity to construct the Pareto front. Although the MOPSO shows good performance on the low dimensional multi-objective optimization problems, it shows poor performance for the high dimensional ones. To overcome this problem, we have already improved the MOPSO using the determination of the initial positions of the particles and the effective Gbest selection method. In addition, we clarified that the improved method shows better performance than the MOPSO for the benchmark problems. However, evaluations of our proposed method for the real-world problems such as the hybrid renewable energy system is still remaining work. Thus,we evaluate the proposed method using the hybrid renewable energy system in this paper to investigate the applicable possibility of our proposed method for the real-world optimization problems.