Abstract: Aiming at the problem that searching precision of Particle Swarm Optimization(PSO) is low and optimized performance is not well for high-dimension function, this paper proposes an improved PSO algorithm. The algorithm uses an orderliness increasing mode to set particle free, enhances the useful population information, leads extreme change through release particle to strengthen computational efficiency of algorithm. Experimental results show that improved algorithm has more powerful optimizing ability and higher optimizing precision compared with other algorithms.

Key words: Particle Swarm Optimization(PSO), large-scale function optimization, release particle, extreme change

摘要： 针对粒子群优化算法搜索精度不高、对高维函数优化性能不佳的问题，提出一种改进的粒子群优化算法。以递增方式对粒子进行释放增强可利用的种群信息，通过释放粒子引导极值变化加强算法的运算效率。实验结果表明，与其他算法相比，改进算法具有更强的寻优能力和搜索精度，且适于高维复杂函数的优化。

关键词: 粒子群优化, 大规模函数优化, 释放粒子, 极值变化

CLC Number: 

• TP301.6