基于改进粒子群算法的时变系统辨识

Identification of Time-Varying Systems Based on Improved Particle Swarm Optimization Algorithm

针对时变系统辨识问题,在基本粒子群算法的基础上,提出了一种改进的粒子群算法对时变系统进行辨识的方法,实质是把辨识问题转换为参数空间上的函数优化问题,然后利用改进粒子群算法的并行搜索能力进行参数寻优。仿真实验表明,与其他算法相比,利用改进算法不仅提高了算法的辨识精度,改善了对时变参数的跟踪精度。

Aiming to identification the time-varying system,on the basis of the basic particIe swarm aIgorithm,an improved particIe swarm optimization method for identification of time-varying system is presented in this paper.The essence is the identifica-tion probIem is converted to the function optimization probIems in parameter space.Then use the paraI eI search abiIity of improved particIe swarm aIgorithm for parameter optimization.