基于K-means聚类的麻雀搜索算法研究

Study on Sparrow Search Algorithm Based on K-Means Clustering

针对一种最新提出的群体优化算法-麻雀搜索算法在寻优过程中存在着过早收敛且容易陷入局部最优解的缺陷。提出一种基于K-means聚类的麻雀搜索方法(KSSA),在初始化种群阶段进行K-means聚类分化,使群体间沟通效率高,增大容错率,从而提高群体开采能力。最后对10个基准函数进行仿真,实验结果表明,所提算法能够客服麻雀搜索算法易陷入局部最优的缺点,提高算法的搜索精度,收敛速度和稳定性。同时将其应用在SVM参数寻优的问题上,验证了KSSA应用于实际问题的可行性。

The sparrow search algorithm, a newly proposed swarm optimization algorithm, has the defects of premature convergence and is easy to fall into the local optimal solution during the optimization process. In this paper, a sparrow search method(KSSA) based on k-means clustering is proposed, and k-means clustering differentiation is carried out in the initial population stage, which makes the communication efficiency between groups high, increases the fault tolerance rate, and thus improves the mining capacity of the group. Finally, simulation experiments are carried out on 10 benchmark functions. The experimental results show that the proposed algorithm can overcome the disadvantage of the sparrow search algorithm falling into local optimal, and improve the search precision, convergence speed and stability of the algorithm. At the same time, it is applied to the optimization of SVM parameters, and the feasibility of KSSA applied to practical problems is verified.