摘要
粒子群算法在函数优化过程中容易陷入早熟收敛和后期搜索精度不高,为防止此问题,采用了在寻优过程中使部分粒子随机初始化,历史最优粒子之间做交叉变异和群体最优粒子做小范围的随机变异的方法。仿真结果表明,与标准的粒子群算法相比改进后的算法能有效避免陷入局部最优和使收敛精度较高。
To avoid the problem of premature convergence and poor accuracy in later period, reinitialized part of particles during the searching process is adopted. Crossover mutation is used for optimum particles and random variation of group optimal particles is used in small range. The simulation experiment indicates that compared with the standard PSO algorithm, the improved PSO algorithm can avoid the local optimum effectively and has better convergence accuracy.
出处
《电脑与电信》
2011年第11期65-66,共2页
Computer & Telecommunication
关键词
粒子群优化
交叉变异
函数优化
particle swarm optimization
crossover mutation
function optimization
