The Improvement on Controlling Exploration and Exploitation of Firework Algorithm

摘要：

　　Firework algorithm (FWA) is a new Swarm Intelligence (SI) based optimization technique, which presents a different search manner and simulates the explosion of fireworks to search the optimal solution of problem.Since it was proposed, fireworks algorithm has shown its significance and superiority in dealing with the optimization problems.However, the calculation of number of explosion spark and amplitude of firework explosion of FWA should dynami cally control the exploration and exploitation of searching space with iteration.The mutation operator of FWA needs to generate the search diversity.This paper provides a kind of new method to calculate the number of explosion spark and amplitude of firework explosion.By designing a transfer function, the rank num ber of firework is mapped to scale of the calculation of scope and spark number of firework explosion.A parameter is used to dynamically control the explo ration and exploitation of FWA with iteration going on.In addition, this paper uses a new random mutation operator to control the diversity of FWA search.The modified FWA have improved the performance of original FWA.By experiment conducted by the standard benchmark functions, the performance of improved FWA can match with that of particle swarm optimization (PSO).

关键词： Firework Algorithm Swarm Intelligence Algorithm Exploration and Exploitation PSO

作者: Jianhua Liu Shaoqiu Zheng Ying Tan

作者单位: School of Information Science and Engineering, Fujian University of Technology,Fuzhou, 350108, P.R.C Department of Machine Intelligence, School of EECS, Peking University Key Laboratory of Machine Perc

会议类型: 国际会议

会议名称: 4th international Conference,ICSI2013(第4届群体智能国际会议)

会议地点: 哈尔滨

会议语种:英文

页码: 11-23

在线出版日期: 2013-06-12（万方平台首次上网日期，不代表论文的发表时间）