粒子群优化算法中的位置矢量的评价策略

为了解决粒子群优化算法(particle swarm optimization,简称PSO)在解决高维多极值问题时容易陷入局部最优而早熟及位置矢量的评价策略存在的"两进一退"和"两退一进"的问题,提出了一种新的评价策略,对各粒子的位置矢量逐维进行评价,使粒子向目标最优位置"稳步前进",具有和标准PSO一样的收敛性分析过程,没有增加对PSO的理解难度,并定义了广义评价策略,实验证明,可以有效地在收敛速度和防止早熟之间平衡以达到很好的优化性能。

The Strategy to Evaluate Position Vector in Particle Swarm Optimization

The particle swarm optimization(PSO) may be trapped in local optima and fail to converge to global optima,especially for multimodal and high-dimensional problems,and it has two shortcomings of "two steps forward,and one step back" and "two steps back,and one step forward".To solve the above problems,a novel evaluation strategy of a particle's position was presented,which has the same convergence analysis as the standard PSO.Using the new evaluation strategy,each particle was evaluated in dimension-by-dimension order so as to step steadily toward the aimed position.Then,a general evaluation strategy was defined.Experiments showed that it could balance effectively the convergence speed and the convergence precision.