摘要
针对标准粒子群算法存在的收敛性和收敛速度的问题,提出一种基于纠错机制的粒子群优化(MPSO)算法。该算法通过对粒子速度的更新过程引入一种简单的纠错机制,使得粒子在进化过程的每一步可能出现的错误得以及时修正,从根本上降低粒子在搜索过程中出错的概率。采用3个典型的函数进行测试,仿真结果表明:与标准粒子群算法相比,该算法有效地提高了其全局收敛能力和收敛速度。
Aiming at problem of convergence and the convergent speed of the standard panicle swarm algorithm, present a panicle swarm optimization algorithm based on error correction mecharlism called MPSO. A simple error correction mechanism is introduced into the update process of the particle, so that the possible error of the particle in every step of the evolutionary process can be corrected in time, and error probability of the particle in searching process can be reduced fundamentally. Using three typical functions to test and simulation results show that global convergence ability and convergence rate of the algorithm are increased effectively compared with the standard particle swarm algorithm.
出处
《传感器与微系统》
CSCD
北大核心
2014年第5期118-120,共3页
Transducer and Microsystem Technologies
基金
山西省自然科学基金资助项目(2012011013-5)
关键词
粒子群优化算法
纠错机制
倒退现象
函数优化
收敛性
particle swarm optimization （ PSO ） algorithm
error correction mechanism
setback
functionoptimization
convergence
