The Analysis And Research Of Iterative Control Algorithm For Nonlinear System

As we all know, nonlinear time-varying systems is extremely difficult to control, even though using modern control theory it is difficult to design an appropriate controller. Iterative learning control is simple and effective for the system of repeatedly tracking the same trajectory , such as operation of Manipulator and the Disk driver system, and so on, Iterative learning control algorithm can realize the complete tracking task in the limited time interval using the information before and the control signals are corrected through iterative approach. iterative learning control algorithm is simple and effective. And this is why the researchers pay more attention to it. the main content includes the composition of iterative learning law, Convergence, Robust, initial values and the learning speed and so on.This paper analyses the iterative learning control's convergence, robustness, initial values and the integration with other advanced control methods Based on the status quo of the iterative learning control. Its main contents are as follows:Firstly the the P-type , D-type and PID-type iterative learning control's convergence and robustness are researched and the simulation based on the different learning laws is also researched. Secondly a method estimating the iterative learning gain is put forward based on the P-type closed-loop iterative control algorithm with the principle of Lyapunov. This method is simple and effective to get the gain using the LMI toolbox in MATLAB. thirdly the method of the integration of iterative learning control and neural network control is researched. This method can fit the parameters of the iterative learning control using the neural network. It gives full play to advantages of the two methods and not only increases the robustness also ensure the accuracy. Lastly the paper analyses the convergence and the robustness of the iterative learning control with the initial states and the random initial states.