A triple-spark guiding strategy to enhance the loser-out tournament-based fireworks algorithm

作     者：Li, Sicheng Zhu, Junhao Han, Mingzhang Fan, Mingjie Zhao, Xinchao 

作者机构：Beijing Univ Posts & Telecommun Sch Sci Beijing 100876 Peoples R China 

出 版 物：《MATHEMATICAL BIOSCIENCES AND ENGINEERING》 (Math. Biosci. Eng.)

年 卷 期：2023年第20卷第4期

页      面：7234-7252页

核心收录：

学科分类：0710[理学-生物学] 07[理学] 0701[理学-数学] 070101[理学-基础数学] 

基　　金：National Natural Science Foundation of China Beijing Natural Science Foundation BUPT Excellent Ph.D Students Foundation [CX2022157] 

主　　题：single-object optimization fireworks algorithm loser-out tournament differential evolution guiding strategy 

摘      要：Loser-out tournament-based fireworks algorithm (LoTFWA) is a new baseline of fireworks algorithm (FWA). However, its ability to search deeply in local areas and communicate among fireworks is not satisfying enough. Therefore, this paper proposes a new triple-spark guiding strategy for LoTFWA to deal with the mentioned problems. Among the three sparks generated for guiding, one is exactly the same as the original one in LoTFWA, the second one uses the centroid of good sparks to enhance the local exploitation, and the last one is based on Differential evolution (DE) mutation and used to enhance cooperation and exploration. Experimental results show that with low computational cost, the proposed guiding strategy attains significantly better results than LoTFWA and is competitive with state-of-the-art FWA variants. Furthermore, comprehensive experiments show that the proposed strategy has the potential to combine with other FWA variants to achieve better results.