摘要
提出一种改进蚁狮优化算法,引入混沌序列进行初始值的分配,增强种群的均匀性和遍历性;在个体更新部分引入粒子群算法的思想,分别以当前的最优个体与全局最优个体为目标进行计算,同时提高算法的局部和全局搜索能力;参考当前最优个体位移进行动态空间收缩,可有效减小个体的搜索范围,缩短寻优时间。与粒子群算法、蝙蝠和原蚁狮算法进行仿真对比并应用到太阳电池模型参数辨识中,验证其有效性。
An improved ant lion algorithm was proposed,which allocates the initial positions of individuals by chaotic sequence,enhancing the population uniformity and ergodicity.The idea of particle swarm algorithm is introduced in the position updating of individuals,and the position of individuals is calculated based on the current best individuals and the overall best individual to enhance the capability of local and overall searching.The dynamic convergence by referring to the current optimal individual displacement is used to decrease the search range and shorten the time of optimization efficiently.The improved ant lion algorithm is compared with particle swarm algorithm,bat algorithm and ant lion algorithm in the identification of parameters of the solar cell model to verify validity.
作者
吴忠强
于丹琦
康晓华
Wu Zhongqiang;Yu Danqi;Kang Xiaohua(Key Lab of Industrial Computer Control Engineering of Hebei Province,Yanshan University,Qinhuangdao 066004,China)
出处
《太阳能学报》
EI
CAS
CSCD
北大核心
2019年第12期3435-3443,共9页
Acta Energiae Solaris Sinica
基金
河北省自然科学基金（F2016203006）
