带非均匀动态变异的改进粒子群优化方法

Improved Particle Swarm Optimization Method with Non-uniformly Dynamic Mutation

针对基本粒子群(PSO)算法在前期收敛速度较快和搜索精度差的缺陷,提出了一种带非均匀动态变异的改进粒子群优化方法。该方法通过引入非均匀动态变异算子不但克服了粒子群算法在后期易陷入局部最优的缺陷,而且极大地增强了群体的多样性,进而提高了算法的搜索效率。最后,通过两个复杂多峰函数的计算仿真,其结果表明该方法是非常有效的。

In order to improve the defects of the simple particle swarm optimization （PSO） algorithm such as premature convergence and rough result,a new particle swarm optimization method with the non-uniformly dynamic mutation operator is presented. By introducing the non-uniformly dynamic operator,the new approach not only can overcome the defaults in which the simple PSO algorithm can easily plunge into the local optimal but also increase the diversity of swarm,and the searching effectiveness of the proposed method is improved. At last,the numerical experiments on two complex multimodal functions show that the improved particle swarm optimization method is effective.