摘要
提出一种改进的克隆选择算法用于解决混沌系统的参数辨识问题.该算法利用抗体的高频变异和受体编辑两种机制有效平衡算法的全局探索与局部开发,并引入向精英抗体学习策略进一步提高算法的收敛质量.对10个优化问题的实验表明:所提出算法在求解精度、收敛速度以及稳定性方面具有更好的性能.以参数未知统一混沌系统的同步控制为研究对象,合理设计同步控制器,并对同步系统的稳定性进行理论分析.通过对同步比例因子的设置,实现统一混沌系统的完全同步、反同步、投影同步等多种同步方式.仿真实验结果表明该方法能够实现对未知系统参数的精确辨识以及驱动-响应系统的有效同步控制,验证了所提方法的可行性与有效性.
An improved clonal selection algorithm was proposed for parameter identification of chaotic systems. Global exploration and local exploration of the algorithm are effectively balanced by two mechanisms of hypermutation and receptor editing. Convergence quality of the algorithm is improved by introducing strategy of learning from elite antibody. Experiments on 10 optimization problems show that the algorithm has good performance in terms of accuracy, convergence speed and stability. Taking synchronization control of unified chaotic system with unknown parameter as research object, synchronization controller was designed reasonably and stability of the synchronization system was analyzed theoretically. By setting synchronization scale factors, a variety of synchronization methods such as complete synchronization, anti-synchronization and projective synchronization of unified chaotic system are realized. Simulation results show that the method realizes accurate identification of unknown system parameter and effective synchronization control of drive-response system. Feasibility and effectiveness of the method are verified.
作者
石建平
李培生
刘国平
SHI Jianping;LI Peisheng;LIU Guoping(School of Mechanical&Electrical Engineering,Nanchang University,Nanchang,Jiangxi 330031,China;School of Electronic&Communication Engineering,Guiyang University,Guiyang,Guizhou 550005,China)
出处
《计算物理》
EI
CSCD
北大核心
2020年第2期240-252,共13页
Chinese Journal of Computational Physics
基金
国家自然科学基金(51566012)
贵州省联合基金(黔科合LH字[2015]7302号)资助项目。
关键词
克隆选择算法
统一混沌系统
混沌同步
参数辨识
clonal selection algorithm
unified chaotic system
chaotic synchronization
parameter identification
