种群规模自适应控制的中心引力优化

Central Force Optimization with Self-Adaptive Control Strategy of Population Size

在中心引力算法的设计中,较大的种群规模能提高最优解的精度,但会降低个体的搜索空间.针对中心引力算法提出了一种自适应控制种群的中心引力算法,在算法的运行过程中,根据算法的表现使每一代增大或减小种群的规模.将聚类算法和佳点集算法融合到增加\删除算子中,使得算法可以自适应地兼顾有效性和多样性.数值结果表明,新算法在求解精度和收敛速度上不弱于对比算法.

In the central force optimization（ CFO）,a large number of population leads to the accuracy of getting an optimal solution. However,owing a large population size will not be a good idea in the case where the search space is small. This article proposed an self-adaptive population size based CFO（ APCFO）,in which the population pool size either grown or shrunk at every iteration based on the performance status of the algorithm. The clustering algorithm and good point set method are used in the design of increase / decrease operator,which considers the effectiveness and diversity adaptively. Experiment shows that the new algorithm no worse than other classical algorithms.