一类非线性系统的多模型神经网络解耦控制器

针对多变量非线性离散时间系统设计多模型神经网络解耦控制器，在每个平衡点处用一神经网络离线辨识非线性系统的线性部分，利用另一神经网络在线辨识非线性部分，将非线性部分视为可测干扰并采用前馈的方法予以消除，所有平衡点处得到的系统模型汇集起来构成多模型集，在每一采样时刻基于切换指标选出最优模型作为当前模型，并据此设计解耦控制器实现控制，仿真结果表明系统在多个平衡点处仍然可以得到较好的控制效果。

Multiple models neural network decoupling controller for a nonlinear system

A multiple models neural network decoupling controller is designed to control the multivariable nonlinear discrete time system. At each equilibrium point, one neural network is trained offline to identify the linear term of the nonlinear system and the other neural network is trained online to identify the nonlinear term. The nonlinear term of the system is viewed as the measurable disturbance and eliminated using feedforward strategy. The multiple models are composed of all models, which are got from all equilibrium points. According to the switching index, the best model is selected as the system model and the decoupling controller is designed accordingly. The simulation (example) shows that a better system response can be got even when the system is changed in many equilibrium points.