基于多元模糊规划的微粒群优化算法

Multi-optimum Fuzzy Programming Based Particle Swarm Optimization

在微粒群的静态多元规划模式的基础上,考虑到多元最优值对群体寻优的引导因子间的比例在寻优过程中不能进行动态自适应调整,因而将模糊逻辑引入对微粒群的多元规划引导,提出了一种用于自适应动态规划的模糊微粒群算法模式,并以最优和次最优分布信息的模糊规划为例,进行了微粒群多元模糊规划模式的设计和数值仿真.仿真结果表明,该算法模式较静态多元规划模式具有更好的总体收敛性能.

In particle swarm optimization, although falling into local optimums can be avoided by introducing the information of multi-optimum distribution state into the particle swarm movement, the performance of the algorithm is limited because the programming proportion factor of multi-optimum cannot be dynamically adjusted in the optimization process. A kind of fuzzy strategy based on double-variable and single-dimensional fuzzy control structure is proposed and used to the dynamic programming of particle swarm multi-optimum. Simulation results show that this kind of fuzzy multi-optimum programming mode has better general convergence performance than traditional PSO algorithm and the static multi-optimum programming mode.