基于RBFNN的非线性动态系统辨识与控制  

The Non - Linear Dynamic System Recognition and Control Based on RBFNN

作　　者：吴晓汉[1] 王金春[1] 王骥程[1] 

机构地区：[1]浙江大学,杭州310027

出　　处：《自动化仪表》1998年第2期24-27,共4页Process Automation Instrumentation

摘　　要：提出了将径向基函数神经网络(RBFNN)的辨识能力与非线性内模控制结合起来得到一种新的控制策略.介绍通过训练RBFNN来得到过程的逆模,由此逆模控制器与一个鲁棒滤波器组成非线性内模控制器.给出了pH中和反应过程的仿真.A new control strategy is stated. It is composed of the recognition capability of the Radial Basic Function Neuro - Network (RBFNN) and the non - linear internal mode control. The inverse model of the process is derived by training the RBFNN and a non -linear internal mode controller is composed by this inverse model and a robust filter. The simulation of pH neutralising reaction process is given.

关 键 词：基函数 神经网络 内模控制 非线性控制 辨识 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP13[自动化与计算机技术—控制科学与工程]