Preaching-inspired swarm intelligence algorithm and its applications

• Published in: Knowledge-based systems Vol. 211; p. 106552
• Main Authors: Wei, Dong, Wang, Zhongbin, Si, Lei, Tan, Chao
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 09.01.2021; Elsevier Science Ltd
• Subjects: Accuracy; Algorithms; Convergence; Heuristic methods; Image segmentation; Image threshold segmentation; Meta-heuristic algorithm; Optimization; Optimization algorithms; Parameter sensitivity; Performance evaluation; Swarm intelligence; Typical engineering applications; Swarm intelligence; Meta-heuristic algorithm; Image threshold segmentation; Typical engineering applications
• ISSN: 0950-7051; 1872-7409
• DOI: 10.1016/j.knosys.2020.106552

Swarm intelligence algorithms have been widely used in both research and engineering fields, but they face the problems of low accuracy and premature convergence, which limit their further applications. Inspired by the preachers’ social behaviors, a novel meta-heuristic swarm intelligence algorithm, Preaching Optimization Algorithm, is proposed in this paper. Its convergence accuracy is effectively improved by improving the initial range of offspring individuals. Meanwhile, by introducing the combined weight including individual fitness and position relationship between individuals, the diversity of individuals is improved, thus reducing the possibility of algorithm premature convergence. In this paper, the parameter sensitivity of the Preaching Optimization Algorithm is analyzed firstly. Secondly, the proposed algorithm is evaluated by comparing it with the other meta-heuristic algorithms on CEC’17 benchmark functions. The results indicate the proposed algorithm has strong competitiveness both accuracy and robustness in solving optimization problems. Finally, the Preaching Optimization Algorithm is used to solve the typical problems in engineering and image threshold segmentation, which further verifies the excellent optimization performance of the proposed algorithm. In this paper, the Preaching Optimization Algorithm is explained in detail and compared with other existing methods to evaluate its comprehensive performance.