一类线性时变系统组合自适应迭代学习辨识

针对一类在有限时间区间上可重复运行的既含时变参数又含时不变参数的高阶线性时变系统,提出了一种模型参考组合自适应迭代学习参数辨识算法.应用Lyapunov方法,给出了时不变参数的时域自适应学习律和时变参数的迭代域自适应学习律,分析了参数估计和模型状态跟踪误差的有界性与收敛性.该算法适于时变和时不变参数并存的线性系统的参数辨识,可加快参数估计的收敛速度.仿真例子验证了所提出的辨识算法的有效性.

Composite adaptive iterative learning identification for a class of linear time-varying systems

For a class of high order linear systems with time-varying and time-invariant parameters which are repeatable in a finite time interval, a composite adaptive iterative learning identification algorithm based on reference model is proposed. By using of Lyapunov technique, a time-domain based adaptive learning law is designed for time-invariant parameters, and an iteration-domain based adaptive learning law is designed for time-varying parameters. The boundedness and convergence of the estimated parameters an...