摘要
针对一类不确定系统,在系统上界值未知的情况下,结合神经网络能任意的逼近不确定系统的优点,设计出一种神经网络积分变结构控制器,利用RBF(Radial Basis Function)神经网络来实时估计系统的不确定性界限,从而降低了一般变结构控制研究的条件。在变结构控制器中又引入饱和函数取代符号函数,进一步减弱"抖振"现象。仿真效果表明,该方法是有效的。
In the case that unknown the upper bound valve for a class of uncertain system, a neural network integral variable structure controller is designed it combines with the advantages of neural network that can approximate any uncertain system. By using RBFNN （ radial basis function neural network ） to real-time estimate the upper bound of the uncertainty system, it reduces the general research conditions of the variable structure control. Symbols function is replaced by saturated function in variable structure control. It can weaken further ＂chattering＂ phenomenon. The simulation results show that the method is effective.
出处
《自动化技术与应用》
2011年第11期1-3,25,共4页
Techniques of Automation and Applications
