摘要: 为完善布谷鸟搜索(CS)算法的收敛性理论,建立CS算法的Markov链模型,分析该Markov链的有限齐次性,在此基础上通过分析鸟窝位置的群体状态转移过程,指出随机序列将进入最优状态集,同时证明CS算法满足随机搜索算法全局收敛的2个条件。通过仿真实验验证CS算法可收敛于全局最优,从而确保CS算法的全局收敛性。
关键词:
启发式算法,
布谷鸟搜索,
Markov链,
状态转移,
全局收敛性
Abstract: In order to perfect the convergence theory of Cuckoo Search(CS) algorithm, the Markov chain model of the CS algorithm is established and the property of the limited and homogeneous of Markov chain is analyzed. On the basis of this, through the analysis of the state transition process of a group of nest position, the stochastic sequence enters to the optimal state set. And CS algorithm meets the global convergence qualification of random search algorithms. Simulation experimental results show that CS algorithm achieves the global optimization, and the global convergence is ensured.
Key words:
heuristic algorithm,
Cuckoo Search(CS),
Markov chain,
state transition,
global convergence
中图分类号:
王凡, 贺兴时, 王燕, 杨松铭. 基于CS算法的Markov模型及收敛性分析[J]. 计算机工程, 2012, 38(11): 180-182,185.
WANG Fan, HE Xin-Shi, WANG Yan, YANG Song-Ming. Markov Model and Convergence Analysis Based on Cuckoo Search Algorithm[J]. Computer Engineering, 2012, 38(11): 180-182,185.