应用随机微粒群算法辨识Hammerstein模型  

Hammerstein model identification based on Random Particle Swarm Optimization algorithm

作　　者：文莹[1] 肖明清[1] 胡雷刚[1] 

机构地区：[1]空军工程大学自动测试系统实验室,陕西西安710038

出　　处：《微计算机信息》2010年第19期180-182,共3页Control & Automation

摘　　要：Hammerstein模型的辨识可以描述成参数空间的函数最优化问题,因此可采用微粒群算法(PSO)对参数并行寻优。针对基本PSO算法容易局部收敛的不足,引入随机操作以提高算法的全局搜索能力。随机微粒群算法(RPSO)在一定条件下能保证全局收敛,用于辨识Hammerstein模型能获得较高精度。仿真结果验证了该方法的可行性和有效性。he system identification of Hammerstein model can be described as function optimization over parameter space,and the Particle Swarm Optimization algorithm can be adopted to obtain optimal solution by searching in parallel. Coping with the basic Particle Swarm Optimization algorithm being easy to run into local optima,random operation was put forward to improve the global searching ability. Random Particle Swarm Optimization algorithm promises global convergence in given preconditions,and can obtain better precision when used to identify the Hammerstein model. The simulation results show the effectiveness of the improved method.

关 键 词：系统辨识 HAMMERSTEIN模型 微粒群算法 非线性系统 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]