一种改进的变异粒子群算法研究

An Improved Particle Swarm Optimizer

粒子群优化算法PSO（Particle Swarm Optimization）是一类性能优越的寻优算法。但由于早熟问题，影响了算法性能的发挥。针对这一问题，引入粒子距离的概念，提出一种新的PSO改进方法（称为NA-PSO）。通过求解组网雷达的系统偏差，表明了NA-PSO算法的可行性，与对比方法相比较，能有效配准，且具有更好的收敛精度和更快的进化速度。

Particle Swarm Optimizer （PSO） is a probability algorithm with excellent performance. But PSO is also of premature phenomenon which limits the effect of PSO. The concept of particle distance is introduced, and an improved method （named NA- PSO） is put forward in this paper. Its application in radar - network proves that NA - PSO is useful and effective, compared with the contrast methods, NA -PSO has better global optimal capability with faster evolution speed and higher accuracy.