Enhanced chicken swarm optimisation for function optimisation problem. (2018)

• Record Type: Journal Article
• Authors: Lin, Min; Zhong, Yiwen; Lin, Juan; Lin, Xiaoyu
• Is Part Of: International journal of wireless and mobile computing. Volume 15:Number 3(2018)
• Page Start: 258
• Page End: 269
• Publication Date: 2018
• Subjects: chicken swarm optimisation -- adaptive neighbourhood -- learning from chicks -- function optimisation -- swarm intelligence; Mobile computing -- Periodicals; Wireless communication systems -- Periodicals; 004.6
• Languages: English
• ISSNs: 1741-1084

Abstract

Chicken Swarm Optimisation (CSO) algorithm is a novel swarm intelligence algorithm. Improper balance between the diversification and intensification may degrade its performance. In order to tackle this problem, this paper proposes an enhanced CSO (ECSO) algorithm which can get better balance between diversification and intensification for the swarm. Specifically, a novel adaptive neighbourhood strategy is used by the location update equation of roosters, so roosters can focus on exploration in early stage and on exploitation in late stage. In addition, learning from chicks is introduced into the location update equation of hens, so that hens can learn from chicks occasionally and increase the diversity of swarm. Experiments on sixteen benchmark problems were conducted to compare the proposed ECSO algorithm with the original CSO algorithm and other classical swarm intelligent algorithms. The results show that ECSO algorithm can achieve good optimisation results in terms of both optimisation accuracy and robustness.