Abstract: The Particle Swarm Optimization（PSO） has problems as being trapped in local minima due to premature convergence and weakness of global search capability. To overcome these disadvantages, Dispersion Particle Swarm Optimization Algorithm based on Random Whip Mechanism（EGPSO） is proposed. Firstly, the concept of particles’ dispersion is presented. In order to avoid falling into local optimum, the algorithm determines the state of the loose particles and marks them by evaluating the dispersion of each particle dynamically, and then uses random whip mechanism to deal with loose particles. Secondly, in order to further improve the algorithm’s convergence speed and accuracy, EGPSO handles active particles by using the optimal location of history. Experimental results on eleven standard benchmark functions demonstrate that EGPSO outperforms original PSO and the other related algorithms in terms of the solution quality and the stability.

Key words: Particle Swarm Optimization（PSO）, random whip mechanism, dispersion, optimization accuracy, convergence speed

摘要： 针对标准粒子群算法全局搜索能力差、易陷入早熟等问题，提出了基于随机鞭策机制的散漫度粒子群算法。首先，给出了粒子散漫度概念，通过动态地对各个粒子的散漫程度进行评估，判断粒子状态，并通过随机鞭策机制处理散漫粒子，避免算法陷入局部最优；其次，对积极运动的粒子利用个体历史最优位置进行处理，加快算法收敛速度；对11个标准函数进行测试，并与标准粒子群算法和其他改进算法进行对比，实验结果表明，基于散漫度的快速收敛粒子群算法寻优精度更高，收敛速度更快。

关键词: 粒子群算法（PSO）, 随机鞭策机制, 散漫度, 寻优精度, 收敛速度