Abstract:

In order to improve the convergence speed and accuracy of cuckoo search algorithm, an improved cuckoo search algorithm based on self-adaptive mechanism is proposed. The improved cuckoo search algorithm uses two different self-adaptive strategies at the beginning and end of iteration to adjust the step size and the discovery probability dynamically, aiming at improving the local and global optimization ability of the algorithm. Ten standard test functions are used to compare the performances of basic cuckoo search algorithm, the improved cuckoo search algorithm and other intelligent optimization methods. The results show that the improved cuckoo search algorithm has certain advantages in solution accuracy, stability and convergence speed.

Key words: computational intelligence, cuckoo search algorithm, self-adaptive strategy, global optimization

摘要：

为提高布谷鸟算法的收敛速度和求解精度，提出了一种基于自适应机制的改进布谷鸟算法。该算法在迭代初期和末期分别使用两种自适应策略来动态调整步长和发现概率，提高了算法的局部和全局寻优能力。利用10个标准测试函数对基本布谷鸟算法、所提出的改进算法以及其他智能优化方法进行了仿真对比验证，结果表明所提出的改进布谷鸟算法在求解精度、稳定性以及收敛速度上都具有一定优势。

关键词: 计算智能, 布谷鸟算法, 自适应策略, 全局寻优