基于神经网络和多模型的非线性离散自适应控制

Nonlinear discrete time adaptive controller based on multiple models and neural networks

针对一类非线性离散时间单变量系统,提出了基于多模型切换策略的非线性自适应控制方法。首先将被控系统划分为多个工作区间,然后在每个工作区间内建立1个线性自适应控制器和1个非线性神经网络自适应控制器。线性控制器可以保证系统的稳定性,神经网络非线性控制器可以有效的改善系统的暂态性能,采用有效的切换策略可以在保证系统稳定的情况下很好的改善系统的性能。仿真结果验证了所提出方法的有效性。

An adaptive controller based on multiple models and neural networks is proposed for a class of nonlinear discrete time system.First, according to different working points,it is divided into multiple zones,in each zone,one linear adaptive controller and one nonlinear neural network controller is designed.The linear controller is designed to ensure the boundedness of the input and output signals,while the nonlinear controller is introduced to improve the performance of the system.Using the switching scheme,both improved performance and stability are achieved simultaneously.In the end,Simulation results show the effectiveness of the proposed method.