一类非线性系统的神经网络自适应区间观测器设计

本文研究了一类单输入单输出非线性系统的神经网络自适应区间观测器设计问题. 针对由状态和输入所描述的未知非线性函数的界不可测, 现有的区间观测器方法并未有效地处理系统含有参数不确定性的未知非线性函数. 首先, 本文构造两个径向基函数神经网络来逼近未知非线性部分, 进而分别估计系统状态的上下界; 然后, 选择合适的Lyapunov函数, 采用网络权值校正和网络误差选择机制确保所设计的误差动态系统有界和非负性, 并证明了神经网络自适应区间观测器的稳定性; 最后, 通过仿真实例验证了所提出的神经网络自适应区间观测器的有效性.

A neural network adaptive interval observer design for a class of nonlinear systems

The problem in designing a neural network adaptive interval observer for a class of single-input single-output nonlinear systems is considered in this paper. The bounds of unknown nonlinear functions described by the state and the input are unmeasurable, so that the existing interval observers are not effective in dealing with unknown nonlinear functions with parameter uncertainty in their systems. In this work, two radial basis function (RBF) neural networks are constructed to approximate the unknown nonlinear part, and then the upper and lower bounds of the system state are estimated, respectively. After chosen a suitable Lyapunov function, network weight correction and network error selection mechanisms are given, which are used to make sure the designed error dynamic system is bounded and non-negative. Furthermore, the stability of the neural network adaptive interval observer is proved. Finally, a numerical simulation example is applied to verify the effectiveness of the proposed neural network adaptive interval observer.