Improving constraint handling for multiobjective particle swarm optimization

Erdong Yu*, Qing Fei, Hongbin Ma, Qingbo Geng

*此作品的通讯作者

科研成果: 书/报告/会议事项章节会议稿件同行评审

4 引用 (Scopus)

摘要

In this paper, a novel particle swarm algorithm for solving constrained multiobjective optimization problems is proposed. The new algorithm is able to utilize valuable information from the infeasible region by intentionally keeping a set of infeasible solutions in each iteration. To enhance the diversity of these preserved infeasible solutions, a modified version of adaptive grid is introduced. In addition, a voting mechanism is designed to balance the preference of infeasible solutions with smaller constraint violation and the exploration of the infeasible region. The effectiveness of the proposed method is validated by simulations on several commonly used benchmark problems. By using the hypervolume indicator, it is shown that the proposed algorithm is more powerful than two other state-of-the-art algorithms.

源语言英语
主期刊名Proceedings of the 33rd Chinese Control Conference, CCC 2014
编辑Shengyuan Xu, Qianchuan Zhao
出版商IEEE Computer Society
8622-8627
页数6
ISBN(电子版)9789881563842
DOI
出版状态已出版 - 11 9月 2014
活动Proceedings of the 33rd Chinese Control Conference, CCC 2014 - Nanjing, 中国
期限: 28 7月 201430 7月 2014

出版系列

姓名Proceedings of the 33rd Chinese Control Conference, CCC 2014
ISSN(印刷版)1934-1768
ISSN(电子版)2161-2927

会议

会议Proceedings of the 33rd Chinese Control Conference, CCC 2014
国家/地区中国
Nanjing
时期28/07/1430/07/14

指纹

探究 'Improving constraint handling for multiobjective particle swarm optimization' 的科研主题。它们共同构成独一无二的指纹。

引用此