期刊
SOFT COMPUTING
卷 24, 期 2, 页码 1185-1211出版社
SPRINGER
DOI: 10.1007/s00500-019-03953-0
关键词
Fireworks algorithm; Updating information; Core firework; Swarm intelligence algorithm; Evolutionary computing
资金
- National Natural Science Foundation Program of China [61572116, 61572117, 61502089]
- National key Technology R&D Program of the Ministry of Science and Technology [2015BAH09F02]
- Provincial Scientific and Technological Project [2015302002]
- Special Fund for Fundamental Research of Central Universities of Northeastern University [N150408001, N150404009]
As a new variant of swarm intelligence algorithm, fireworks algorithm (FWA) exhibits promising performance on a wide set of optimization problems, for which the fireworks algorithm has been concentrated on and investigated by researchers recently. This paper aims to improve the performance of the FWA by exploiting updating information of the core firework to guide the algorithm's searching process. Based on this mentality, this paper ameliorated the explosion strategy of core firework of dynamic fireworks algorithm (dynFWA). The proposed algorithm, named dynPgFWA in this paper, improved FWA from two aspects: amplifying the explosion amplitude on the direction on which core firework is updated, and making more sparks which are generated by core firework distributed on this direction to enhance the algorithm's searching ability on updating direction. A numerical experiment on CEC2015 and CEC2017 test suite was implemented to verify the performance of the proposed algorithm. Results of the experiment indicated that dynPgFWA outperformed the compared evolutionary algorithms in the quality of solutions.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据