非线性离散系统的自适应反馈一步向前预测控制方法

针对时变、滞后、反馈时间慢且参数未知的非线性离散时间模型系统的特点，提出了一种自适应反馈控制方法。在方法中，考虑到调整时间内反馈信息少与工程易实现的要求，采用多层预报的方法，对模型跟踪误差进一步向前预测，进而利用预测结果对模型和输入做进一步修正。此外，对算法的收敛性和稳定性从数学上给出了说明。并通过数值仿真验证了方法的可行性和有效性。

Adaptive control method with one-step-ahead prediction of a nonlinear discrete system

On the basis of the characteristics of a nonlinear discrete time model system, such as time variable, delay and unknown parameters, a kind of adaptive feed back control methodology is presented. In the method, a multi layer predictive method with one step ahead prediction is proposed to face, with little feed back information in the required adjusting time, the requirement of easier engineering realization. Moreover, the predictive result is empolyed to improve the model and input parameter. Furthermore, the proof of convergence and stability of the method is developed mathematically. Finally, three examples are given to demonstrate the methodology.