融合PSO算法思想的进化算法

Evolutionary algorithm based on the idea of particle swarm optimization

粒子群算法(particle swarm optimization,PSO)是仿真于生物群体的社会行为的一种智能优化算法,其原始形式难以体现数学的直观性和本质性。然而,在简化算法原始模型的基础上,PSO算法的理论分析得到其数学模型,并且说明了其是一个迭代进化系统。利用PSO算法的数学模型代替标准PSO算法速度及位置的迭代公式,并选择适当的参数,从而构造了一种新的进化算法。新的进化算法形式更能直接体现PSO算法的数学思想。经仿真试验表明,新的进化算法效果不差于标准PSO算法,并且参数少且容易分析。

Particle swarm optimization（PSO）is an intelligence algorithm simulating the social behavior of a bird swarm or fish group.It is difficult for the original formula of PSO to show mathematical essence and principle.Using the simplified modal of PSO,the current theoretical analysis of PSO constructed a mathematical modal giving a clear essence of PSO from a mathematical view,this illustrated that the PSO was an iteration evolutionary system.Using the mathematical modal of PSO,a new evolutionary algorithm in which the velocity and location updating equation of PSO were replaced by the mathematical equations were developed.Some parameters of the new algorithm were discussed and properly selected.With selection of appropriate parameters,the performance of new evolutionary algorithm was not inferior to the standard PSO by simulation on benchmark functions.The new evolutionary algorithm was easy to understand and had mathematical meaning.Its parameters were fewer and easier to be analyzed than the standard PSO.