基于高斯变异的量子粒子群算法  

An Improved Quantum-behaved Particle Swarm Optimization Based on Gaussian Mutation

作　　者：石永生[1,2] 陈家琪[1] 

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093 [2]江苏信息职业技术学院计算机工程系,江苏无锡214101

出　　处：《电脑与信息技术》2010年第6期9-12,共4页Computer and Information Technology

基　　金：中国博士后科学基金(20090460323)

摘　　要：粒子群算法相对于其他优化算法来说有着较强的寻优能力以及收敛速度快等特点,但是在多峰值函数优化中,基本粒子群算法存在着早熟收敛现象。针对粒子群算法易于陷入局部最小的弱点,提出了一种基于高斯变异的量子粒子群算法。该算法使粒子同时具有良好的全局搜索能力以及快速收敛能力。典型函数优化的仿真结果表明,该算法具有寻优能力强、搜索精度高、稳定性好等优点,适合于工程应用中的函数优化问题。Comparing with the other optimization algorithms,Particle swarm optimization（PSO） has comparable or even superior search performance for many optimization problems with faster and more stable convergence rates.But it can＇t guarantee to find the global optima in the search space especially in multimodal function.For conquering the shortcoming of particle swarm optimization,a novel quantum-behaved particle swarm optimization based Gaussian mutation（GQPSO） is introduced in this paper.The improved algorithm not only has more power global searching ability but also does well in local searching.Experiment simulations shows that the proposed algorithm has powerful optimizing ability,good stability and higher optimizing precision,so it can be applied in optimization problems.

关 键 词：粒子群 高斯 变异 全局搜索 收敛速度 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]