基于细致化仿生的改进粒子群优化算法

粒子群优化(particle swarm optimization, PSO）算法基本思想是试图通过模拟鸟群觅食中的迁徙和聚集等行为获得连续非线性函数的最佳值，其仿生算法产生于对鸟群寻食过程中飞行方向与飞行速度等的隐喻。近年对粒子群算法经典算法的研究，虽然在速度及精度上有所改进，但由于缺乏细致化仿生（precise bionic metaphor, PBM），改进效果并不太明显。通过在PSO算法中引入飞鸟寻食细致化行为特征隐喻，即在算法中同时导入满意粒子局地细致化寻优和探索粒子随机寻优过程，进而提出了一种新的基于细致化仿生的改进PSO算法；对改进算法和经典算法进行了性能比较，结果显示所提算法在收敛速度和求解精度方面较经典算法有很大程度的改善。

Improved particle swarm optimization based on precise bionic metaphor

The basic idea of particle swarm optimization(PSO) is to obtain the optimum value of the continuous nonlinear function by simulating birds behavior such as the direction and the speed of the flight in migration and foraging aggregation.In recent years,the results improvement of the classical PSO algorithm are not obvious because of the lack of precise bionic metaphor(PBM).By introducting PBM into the PSO,a new improved PBM-PSO is set up.The results show that the improved particle swarm optimization converges more quickly and gets a more accurate solution than the classical algorithm.