摘要
采用粒子群算法处理约束优化问题时,由于约束条件使得解空间成为非凸集合,粒子容易陷入局部最优,因此在搜索过程的不同阶段,提出变步长因子的粒子群算法,实验证明改进的算法在精度与稳定性上明显优于采用罚函数的粒子群算法和遗传算法等其他一些算法.
An improved partical swarm optimization, different scale factor partical swarm algorithm was proposed. The numerical results showed that the improved PSO was feasible and can get more precise results than particle swarm optimization by using penalty functions and genetic algorithm and other optimization algorithms.
出处
《天津师范大学学报(自然科学版)》
CAS
2006年第2期73-76,共4页
Journal of Tianjin Normal University:Natural Science Edition
基金
天津市教育科学基金项目(20030515)
关键词
粒子群算法
动态罚函数
变步长因子
particle swarm algorithm
dynamic penalty function
different scale gene