摘要
提出了一种分层交换差分粒子群优化算法。作为一种混合算法,将粒子群优化算法与差分进化算法有机地结合了起来,使得粒子种群和差分种群分层交换,共享最优信息,加快了收敛速度,降低了陷入局部最优的风险,函数测试结果表现出算法具有更好的寻优性能。将算法应用到翼型设计之中,取得了良好的优化效果。
A new hybrid algorithm called Rank-Altered Differential Particle Swarm Optimization（RADPSO） is proposed.In this hybrid algorithm,the population is ranked and altered and the best individual information is shared by both PSO population and DE population.This accelerates the convergence speed and reduces the probability of strapping into a local minimal.Function test results show that RADPSO algorithm has better optimal ability compared with PSO algorithm and DE algorithm.In this paper,RADPSO algorithm is also applied to airfoil design and the result prove it has better optimal effects.
出处
《航空计算技术》
2010年第6期66-71,共6页
Aeronautical Computing Technique
关键词
粒子群优化算法
差分进化算法
分层交换差分粒子群优化算法
翼型设计
particle swarm optimization
differential evolution
rank-altered differential particle swarm optimization
airfoil design
