一种改进的烟花算法  

An Improved Firework Algorithm

作　　者：王苏霞 陈得宝[1] 邹锋[1] WANG Su-xia;CHEN De-bao;ZOU Feng(College of Physics and Electronic Information,Huaibei Normal University,Huaibei 235000,China)

机构地区：[1]淮北师范大学物理与电子信息学院,安徽淮北235000

出　　处：《长春师范大学学报》2020年第6期59-68,共10页Journal of Changchun Normal University

基　　金：国家自然科学基金资助项目“集成动态优化及应用研究”(61976101);安徽省自然科学基金项目“离散教学多目标优化及在个性化推荐中的应用研究”(1708085MF140);安徽省高等学校自然科学研究重点项目“面向图像分割的多目标核聚类模型与方法”(KJ2019A0593)。

摘　　要：针对传统烟花算法(fireworks algorithm,FWA)多样性差、易陷入局部搜索的问题,提出一种改进的烟花算法(CFWA)。传统烟花算法最优个体的爆炸半径非常小,导致其生成的火花之间差异很小,多样性受到了很大的限制。本文采用cubic混沌序列确定爆炸半径,避免了个体半径出现接近于零的现象,改善了火花的多样性。在18个典型函数集上对CFWA算法与jDE算法、PSOFDR算法、HBSA算法、LFWA算法以及EFWA算法进行3项指标对比实验,实验结果显示,CFWA算法在收敛速度和收敛精度上占据显著优势。This paper proposes an improved fireworks algorithm for the problems of traditional fireworks algorithm(FWA)with poor diversity and easy to fall into local search.The explosion radius of the optimal individual of the traditional firework algorithm is very small,the difference between the sparks that caused it to be small and the diversity is greatly restricted.In this paper,the cubic chaotic sequence is used to determine the explosion radius,which avoids the phenomenon that the individual radius is close to zero,and improves the diversity of sparks.On the 18 typical function sets,the CFWA algorithm is compared with the jDE algorithm,PSOFDR algorithm,HBSA algorithm,LFWA algorithm,and EFWA algorithm.The experimental results show that the CFWA algorithm occupies a significant advantage in convergence speed and convergence accuracy.

关 键 词：烟花算法 爆炸 变异 映射 选择 混沌 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]