新型混合粒子群优化算法

针对粒子群算法易陷入局部极值、精度低等缺点,提出了一种基于模拟退火与混沌思想的新型粒子群优化算法(SA-CPSO).在该算法的初始阶段,对粒子位置进行混沌初始化,并引入模拟退火算法对每个粒子的适应度进行评价;在该算法运行过程中根据群体适应度方差对粒子群进行混沌更新;最后通过对几种经典函数的测试计算,结果表明,相对于标准粒子群算法,该新型混合算法提高了局部搜索能力和搜索精度,并有效避免了早熟现象的产生.

New Hybird Particle Swarm Optimization Algorithm

To avoid trapping to local minima and improve the searching performance of simple particle swarm optimization(PSO) algorithm,a new hybird particle swarm optimization based on the simulated annealing(SA) and chaos is proposed.At the beginning stage,the location of the particle is initialized by chaos and then evaluates the fitness of particle by SA.During the running time,according to the variance of the population's fitness,the chaotic update of the particle is performed adaptively.At last the experimental results using the testing functions show that it is prior to standard PSO in search time and precision,and avoids the premature convergence.