摘要
布谷鸟搜索算法(Cuckoo Search,CS)是一种新型的元启发式算法。针对CS算法局部搜索能力较弱、后期收敛速度偏慢和收敛精度不高等缺点,提出一种基于变尺度法(DFP)和自适应步长(Adaptive Step)的布谷鸟搜索算法(DACS),使Lévy飞行的步长非线性自适应变化来提高算法的收敛速度,同时使经过Lévy飞行机制和淘汰机制进化后的布谷鸟种群再运用DFP快速获取全局最优解。用6种具有各种代表性的测试函数分别测试DACS算法和CS算法的性能。实验结果表明,DACS算法在保持强大的全局搜索能力的同时,比CS算法具有更快的收敛速度、更高的收敛精度和更好的鲁棒性,尤其适合多峰及高维函数的优化。
Cuckoo Search （CS） is a novel meta-heuristic algorithm. Aiming at the defects of weak local search ability ,slow convergence velocity and low convergence accuracy, a modified CS algorithm based on DFP and adaptive step is proposed in this paper. In the im- proved cuckoo search algorithm, the step of Levy flight nonlinear dynamic changes improve convergence velocity. After evolved from Levy flights and elimination mechanism, the cuckoo populations rapidly access to global minima by DFP. Sixth representative benchmark functions are used to test the performance of DACS algorithm and CS algorithm respectively. The conclusions indicate that DACS algo- rithm has faster convergence speed, higher convergence accuracy and robustness, compared with CS algorithm. Meanwhile, DACS algo- rithm keeps strong global search capability, which is particularly suitable for the optimization of multimodal function and high dimension function.
出处
《计算机技术与发展》
2015年第10期38-43,共6页
Computer Technology and Development
基金
国家基础科学人才培养基金资助项目(J1103303)
关键词
布谷鸟搜索
变尺度法
自适应步长
全局寻优
cuckoo search
variable metric method
adaptive step
global optimization
