基于精英策略改进的自适应布谷鸟搜索算法

Improved adaptive cuckoo search algorithm based on the elite strategy

传统的布谷鸟搜索算法,在二维优化问题中经常出现陷入局部最优、寻优解精度低以及计算解偏差大等问题。针对于此,本文提出了一种基于精英策略改进的自适应布谷鸟搜索算法。该算法仿生生物群体中的精英策略,将布谷鸟群体按照黄金分割,划分为普通布谷鸟与精英布谷鸟两类。普通布谷鸟基于Levy飞行寻找巢穴寄生,提高了寻优解的精度,加快了后期的收敛速度;而精英布谷鸟则会围绕群体中心位置进行定向侦查,为种群提供更优的巢穴,提高了算法的鲁棒性,避免算法过早成熟,加强了布谷鸟群的全局搜索能力。同时,布谷鸟被寄主鸟发现的概率基于迭代次数而自适应变化,保证了布谷鸟后期将不断收敛于全局最优解。通过仿真对比实验,表明本文提出的改进算法在二维全局寻优上具有较好的鲁棒性与准确性。

Traditional cuckoo search algorithm often suffers from local optimum,low precision of searching solution and large deviation of calculating solution in two-dimensional problem optimization.In this paper,an improved adaptive cuckoo search algorithm based on elite strategy is proposed.The algorithm mimics the elite strategy in the biological population and divides the cuckoo population into ordinary cuckoo and elite cuckoo according to the golden section.Simple cuckoo searchs for nest parasites based on Levy flight,which improves the accuracy of the optimal solution and accelerates the convergence rate in the later stage.The elite cuckoo will conduct directional investigation around the center of the population to provide a better nest for the population,which improves the robustness of the algorithm,avoids premature maturity of the algorithm,and strengthens the global search ability of the cuckoo group.At the same time,the probability of cuckoo being discovered by host bird adaptively changes based on the number of iterations,which ensures that cuckoo will converge to the global optimal solution in the later period.The simulation results show that the improved algorithm has better robustness and accuracy in two-dimensional global optimization.