Research And Application Of Iterative Learning Control

Iterative learning control (ILC) is developed accompanying the application of computer technology and is a control technique of new study which is combined with artificial intelligence and automatic control technology. It can improve the tracking performance and transient response of control systems which performs the same tasks in a finite time interval. According to the research actuality, this dissertation aiming at linear time-invariant (LTI) continuous system has studied the problems on the convergence, the robustness of PID-type learning algorithm, the designing optimal parameter of PID controller. Meanwhile, aiming at industrial process control in two categories: one is the control system can accurately be described by the mathematical model and the other is the control system doesn't have any specific mathematical models, has been analyzed and resolved by ILC theory. Through the practical application, it has proved the good performance of this control technology.The main contents and results in this dissertation are as follows:l. The convergence and robustness of PID-type learning algorithm has been researched. The convergence is proved and the convergent conditions are also given about the open-loop and closed-loop PID-type learning algorithm. Further the convergence and robustness of PID-type learning algorithm is proved when there is the initial existence of bias, state disturbances and measurement noise in the ILC control system.2. The methods about the designing optimal parameter of open-loop and closed-loop PID controller based on the linear time-invariant (LTI) continuous system have been discussed. Furthermore, it is thoroughly researched that design is applied to the discrete system about open-loop iterative learning controller PID optimal parameter based on certain performance index. According to conclusions that the method applied to the discrete system, this design method is applied to continuous system within certain limit conditions.3. The problem about DC motor speed tracking based on ILC has been studied. The control demand about the motor speed tracking in electrical control system is discussed and a specific mathematical model about control system is built. The tracking results of the open-loop and closed-loop PID-type ILC solving such problems are discussed. And the results are also discussed when the system has initial error, state disturbances and measurement noise. It is tested that not only does ILC have good performance, but also has strong robustness and stability through a large number of simulation and experiments are done by MATLAB software.4. The practical engineering problems about stabilizing the tank's liquid level based on ILC have been studied. The water-delivery pump is controlled by this control model through changing its given frequency, thus the water draining speed is changed. These questions belong to uniform control system problems. The engineering problems are solved by using ILC theory on the conditions that the control system couldn't establish specific mathematical model. The building process of ILC model is further discussed. At the same time the results are detailed simulated according to the demand. The model is applied on the industry scene under the design methods are successful. Finally, the hardware design and the software design are elaborated in details.5. The learning algorithms and engineering applications in the dissertation have been testified and simulated through a large number of M-file and the Simulink interface program is programmed by MATLAB software.