Abstract: In order to overcome the shortcomings that standard Particle Swarm Optimization（PSO） traps into local optima easily and has a low convergence accuracy，an improved PSO algorithm is proposed.Double exponential probability distribution is utilized to improve the equation of the velocity and dynamically adjust the maximal velocity of the particle，which increases the diversity of the population and the ability of particle to escape from the local optima and ensures the continual convergence during the course of finding global optima.Experimental results on five representative benchmark functions show that the proposed PSO algorithm improves velocity of convergence in the latter phase，avoids premature convergence problem effectively and has a higher convergence accuracy.

Key words: particle swarm optimization, premature convergence, double exponential probability distribution

摘要： 针对标准粒子群算法容易陷入局部最优、收敛精度低的缺点，提出了一种改进的粒子群算法。它用双指数分布改进了速度方程度，并用其动态地调整粒子的最大速度，扩大了群体的多样性，增强了粒子跳出局部最优解的能力，保证了整个寻优过程的持续收敛。通过比较和分析5个典型测试函数的实验结果，改进的粒子群算法提高了迭代后期的收敛速度，有效地避免PSO算法的早熟收敛问题，而且具有较高的收敛精度。

关键词: 粒子群优化, 早熟收敛, 双指数分布