Study On Robust Control Strategy For Iterative Learning Control System In The Iteration-domain

In the system of iterative learning control, the non-repetitive disturbance is ubiquitous. While the traditional algorithms of iterative learning control are difficult to deal with this adverse effect on the system’s performance and the tracking trajectory with baseline error. Therefore it is necessary to do research on the design of robust control along the iteration direction. Mainly focus on the harmful effects made by the non-repetitive disturbance to the system, a new control structure is proposed, which eliminate or restrain the non-repetitive disturbance and make the system has good robustness, at the same time, guarantee the system with a great tracking performance.The current situation and the existing problems of ILC are summed up. Aiming at a kind of industrial object with repeatability called discrete repetitive processes, use a lifting technique and the notion ofω-transform in the iteration-domain and the approach for designing feedback controllers for discrete repetitive processes given by K.L.Moore to design a non-unit memory controller to track the iteration-varying reference signals. The internal model principle is used to ensure perfect tracking along the pass and asymptotically-convergent tracking along the iteration axis. For the system with iteration-varying disturbance, the disturbance suppressors based on adaptive inverse control is considered. In the case the object’s parameters are unknown and iterative changed, the idea of parameter estimation adaptive control is cited, then the method of system identification is used to establish the model and inverse model of the plant offline or online. Build disturbance suppressors and combine it with the controller of adaptive iterative learning based on internal model principle, then a new control structure is proposed, which can ensure the stability and error convergence of system. Add different types of iteration-varying disturbance to illustrate the effectiveness of the new control structure by MATLAB simulation experiments. Finally make the conclusion that the non-repetitive disturbance is effectively eliminated and the system receive a great tracking performance by this new control structure.