作者: Shaoqiu Zheng , Junzhi Li , Andreas Janecek , Ying Tan
DOI: 10.1109/TCBB.2015.2497227
关键词: Differential evolution 、 Convergence (routing) 、 Algorithm design 、 Fireworks 、 Swarm intelligence 、 Particle swarm optimization 、 Mathematical optimization 、 Artificial intelligence 、 Benchmark (computing) 、 Operator (computer programming)
摘要: This paper presents a cooperative framework for fireworks algorithm CoFFWA. A detailed analysis of existing FWA and its recently developed variants has revealed that (i) the current selection strategy drawback contribution firework with best fitness denoted as core overwhelms contributions all other non-core in explosion operator, (ii) Gaussian mutation operator is not effective it designed to be. To overcome these limitations, CoFFWA proposed, which significantly improves exploitation capability by using an independent method also increases exploration incorporating crowdness-avoiding among fireworks. Experimental results on CEC2013 benchmark functions indicate outperforms state-of-the-art variants, artificial bee colony, differential evolution, standard particle swarm optimization SPSO2007/SPSO2011 terms convergence performance.