一类纯反馈非仿射非线性系统的自适应神经网络变结构控制

Adaptive neural network variable structure control for a class of non-afine nonlinear pure-feedback systems

研究了一类非仿射的纯反馈单输入单输出非线性系统。针对此系统,在中值定理、神经网络参数化和解耦Backstepping的基础上,提出了一种自适应变结构神经网络控制策略,而且所给出的定理证明闭环系统的所有信号在平衡点上是半全局一致有界的。通过对一个非仿射CSTR对象的仿真验证了该方法的有效性。

A class of nonlinear non-affine pure-feedback SISO systems with unknown nonlinear functions are investigated. An adaptive variable structure control is presented for this class of systems based on the combination of mean value theorem, neural network parameterization, and decoupled backstepping design. All the signals in the closed-loop system can be shown to be semi-globally uniformly ultimate boundedness around the equilibrium point. The effectiveness of the proposed control law is verified via simulation for a non-affine CSTR plant.