摘要
为提高布谷鸟搜索算法的搜索能力和寻优精度,提出一种具有动态惯性权重的布谷鸟搜索算法。该算法引入动态惯性权重改进鸟窝位置的更新方式,依据动态惯性权重值保留上代鸟窝的最优位置并进行下一代位置更新,从而有效平衡种群探索能力和开发能力之间的关系。并利用特征方程对改进算法进行了收敛性分析。仿真实验结果表明,与基本布谷鸟搜索算法、粒子群算法和蚁群算法相比,改进后的布谷鸟搜索算法能显著减少迭代次数和运行时间,有效提高算法的收敛速度和收敛精度。
In order to improve the search ability and optimization accuracy of cuckoo search algorithm,the cuckoo search with dynamic inertia weight is proposed. By utilizing the dynamic inertia weight,the improved cuckoo search updates the next nest position based on the former best nest position that has been saved with dynamic inertia weight,which can well balance the relation between population exploration and development capabilities. This paper also has a convergence analysis of the improved cuckoo search by the characteristic equation. The performance of the new method is compared with the basic cuckoo search,particle swarm optimization,ant colony optimization and other algorithms,showing that the improved cuckoo search algorithm can significantly reduce the number of iterations and running time,and can effectively improve the convergence speed and convergence precision.
出处
《智能系统学报》
CSCD
北大核心
2015年第4期645-651,共7页
CAAI Transactions on Intelligent Systems
基金
河南省科技攻关重点基金资助项目(122102210201)
河南大学研究生教育综合改革基金资助项目(Y1427056)
关键词
布谷鸟搜索算法
函数优化
莱维飞行
动态惯性权重
种群规模
收敛性
复杂度
参数选取
cuckoo search algorithm
function optimization
Lévy flight
dynamic inertia weight
population size
convergence
complexity
parameter selection
