The Study Of Iterative Learning Control Based On DSPACE

Iterative learning control (ILC) as a new learning control technology is the combination of automatic control technology and artificial intelligence. It can improve the tracking performance and transient response of control systems which performs the same tasks in a finite time interval, such as a robotic manipulator in a manufacturing environment or a chemical reactor in a batch processing application. In the filed of iterative learning control study, both theory and application have made some achievements. However, on the theory of algorithms, for obtaining good tracking performance, most algorithms'structure are complex and their control effects are not satisfactory;In addition, the application of this technology in practical engineering is not mature.For the above-mentioned problems, the main work of this paper is as follows:1. In most cases, there is no reference index for selecting gain parameter in the application of basic iterative learning control law, only depends upon designer's experience to estimate and try, and the tracking error's monotonic convergence cannot be guarantee. In view of these questions, we have given two parameter optimization methods: a kind of adaptive parameter optimization iteration learning control algorithm is proposed for discrete system, the algorithm has guaranteed the parameter is optimal on each operation; An optimal design method is proposed to determine the PID coefficients for control discrete-time SISO LTI systems in order to improve the transient tracking performance. The simulations results indicated that the above two algorithms have the following characters: the algorithms are simple, the convergence rate is quick, and can guarantee the tracking error is monotonic convergence.2. For second-order SISO linear slow time-varying system, a new method called based on iterative learning model reference adaptive control algorithm is proposed, which is the combination of model reference adaptive control and iterative learning control. The algorithm not only realized the learning in the trajectory tracking, and also implemented the study in the system. An illustrative example is given to demonstrate that the algorithm reduced system's adjust time on tracking desired trajectory.3. Introduced the dSPACE real-time simulation system's hardware composition, the software environment, the real-time simulation function as well as the procedure of developing control system based on the dSPACE.4. Based on the dSPACE hardware and software platform, has designed a fast control prototype of the DC servo motor, and implemented the application of iterative learning control in DC servo position control system.