Abstract: The particle swarm optimization algorithm is a kind of intelligent optimization algorithm.This algorithm has some demerits，such as relapsing into local optima，slow convergence velocity and low convergence precision in the late evolutionary.A new algorithm，based on the new inertia weight，is proposed to overcome the demerits of the basic particle swarm optimization.Six benchmark functions are tested and the experimental results show that the new algorithm not only significantly speeds up the convergence，but also effectively solves the premature convergence problem.

Key words: swarm intelligence, evolutionary computation, particle swarm optimization, inertia weight

摘要： 为了改进基本粒子群算法的搜索功能，针对粒子群算法易于陷入局部极值，进化后期的收敛速度慢和精度低等缺点，通过公式分析得到新的惯性权重调节方法，提出了一种新的改进粒子群算法。用几个经典测试函数进行实验，实验结果表明，新算法不仅具有更好的收敛精度，而且能更有效地进行全局搜索。

关键词: 群体智能, 进化计算, 粒子群算法, 惯性权重