粒子群算法的改进与比较研究

Improvement and comparison of particle swarm optimization

针对粒子群算法(particle swarm optimization,PSO)"早熟收敛"和后期收敛速度慢的特点,文章提出了一种改进的PSO算法。该算法摒弃了近年来许多在改进过程中引入过量繁琐公式、各种变换因子而导致算法过程冗杂的粒子群改进方法,而是在简化PSO算法的基础上引入自适应局部搜索因子,在种群不变的情况下拓宽了搜索范围并提高了搜索精度,且在某些测试函数下寻优效果明显优于其他复杂的PSO优化算法。最后的测试实验表明,该文算法能避免早熟问题,有效地提高了算法的精确寻优能力。

Aiming at the premature convergence and the slow convergence speed in late stage of particle swarm optimization(PSO),an improved PSO algorithm is proposed.Many improved PSO algorithms in recent years introduce excessive fussy formulas or various transformation factors in the improvement process,which makes the algorithm process too complex. Unlike these algorithms,the proposed algorithm introduces adaptive local search factor on the basis of simplified PSO algorithm,which broadens the search scope and improves the search accuracy under the condition of constant population number.Under some test functions,the proposed algorithm is obviously better than the other complex PSO algorithms in recent years.Finally,the experimental results show that the proposed algorithm can avoid premature problems and effectively improve the accuracy of the algorithm.