摘要
针对烟花算法后期收敛速度慢,易陷入局部最优等问题,本文通过引入μ律特性曲线将适应度值映射为适应度等级重新定义爆炸算子,动态调整μ的值以及缩放因子F,更好的平衡了算法的局部开采和全局搜索能力,同时借助基于信息交流的映射策略,增强映射后火花的搜索效率,提高了算法的收敛速度.文中选用12个通用的基准测试函数对μFWA算法与其它四种算法进行对比,实验结果表明,μFWA算法的单峰函数和多峰函数的寻优结果理想,寻优效率高,全局搜索和局部开采能力更加平衡.
Simple firework algorithms get local optimal solutions easily and converge slowly,this paper adjusts the algorithm accordingly.Firstly,redefine the explosion operator by the aid ofμ-law where the fitness value is converted to the degree of fitness,and the aberrance scope can be adjusted by the value ofμand the scaling factor.At the same time,the mapping strategy based on information exchange is used to enhance the search efficiency.In this paper,12 general benchmark functions are used to compare theμFWA algorithm with other four algorithms.The experimental results show that theμFWA algorithm can obtain ideal results in the optimization of unimodal function and multimodal function which means that the algorithm can balance global search and local search well,and the convergence speed of the algorithm is improved obviously.
作者
滕志军
李哲
王幸幸
皇甫泽南
申博冉
TENG Zhi-jun;LI Zhe;WANG Xing-xing;HUANGFU Ze-nan;SHEN Bo-ran(Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology, Ministry of Education, Northeast Electric Power University, Jilin 132012, China;School of Electrical Engineering, Northeast Electric Power University, Jilin 132012, China;College of Mathematics, Jilin University, Changchun 130022, China)
出处
《陕西科技大学学报》
北大核心
2022年第2期187-194,共8页
Journal of Shaanxi University of Science & Technology
基金
国家自然科学基金项目(61501107)
吉林省教育厅“十三五”科学研究规划项目(JJKH20180439KJ)。
关键词
烟花算法
μ律特性曲线
爆炸算子
缩放因子
映射策略
fireworks algorithm
μ-law characteristic curve
explosion operator
scaling factor
mapping strategy