摘要
作为一种基于群智能的并行随机优化方法,粒子群优化算法(PSO)在优化求解问题中体现出了良好的性能.从提出至今引起了广泛的关注,研究成果也不断涌现.从2000年开始,PSO被用于动态优化问题中.这对PSO的研究提出了新的挑战,对于动态问题的优化不再是在解空间中找到一个最优点,而是要尽可能地在解空间中跟踪运动变化的最优点.对目前为止对于PSO在动态环境优化问题的研究内容进行了分析和总结,介绍了针对动态环境优化问题PSO的改进方法、对环境变化的检测和应对策略、优化性能评价的一系列方法以及各种试验及应用案例.
Particle swarm optimization （PSO）, a parallel random optimization method based on swarm intelligence, exhibits good performance for optimization problems. Since 2000, PSO has been applied to optimization problems in dynamic environments. The challenge with PSO is that the objective is not only to locate an optimum, but also to track that moving optimum as closely as possible. This paper presented the latest developments of PSO in dynamic environments. Various research approaches were reviewed, including improvements in PSO, dynamic change detection, response strategies, performance evaluation and experiments used in researching dynamic problems.
出处
《智能系统学报》
2009年第3期189-198,共10页
CAAI Transactions on Intelligent Systems
基金
高等学校优秀青年教师教学科研奖励计划资助项目(20010248)
关键词
粒子群优化方法
动态环境优化
检测策略
应对策略
性能评价
particle swarm optimization （PSO）
optimization in dynamic environment
detection strategy
response strategy
performance evaluation
