PSOSA混合优化策略

PSOSA Hybrid Optimization Strategy

本文提出了一种微粒群算法与模拟退火算法相结合的混合优化方法,该方法在群体进化的每一代中,首先通过微粒群算法的进化方法来控制微粒的飞行方向,然后利用模拟退火算法来拓展其搜索领域。这样既可以利用微粒群算法的收敛快速性,又可以利用模拟退火算法的全局收敛性。本文还证明了该混合优化方法依概率1收敛于全局最优解。仿真结果表明,在搜索空间维数增大时,该方法的全局收敛性明显优于基本微粒群算法。

This paper proposes a hybrid optimization strategy based on particle swarm optimization and the simulated annealing algorithm. This novel method control the direction of particles by particle swarm optimization first, then it uses the simulated annealing algorithm to search in greater search area. The hybrid method utilizes the rapid convergence of particle swarm optimization and the global convergence of the simulated annealing algorithm. This paper also proves that this new algorithm is a global optimization algorithm. Simulation results show that the global comvergence of the novel algorithm is superior to particle swarm when the search dimension is large.